IFAC Journal of Systems and Control最新文献

{"title":"Data-driven modeling with prior system knowledge","authors":"Fritz A. Engeln ,&nbsp;Jan-Willem van Wingerden ,&nbsp;Timm Faulwasser","doi":"10.1016/j.ifacsc.2026.100384","DOIUrl":"10.1016/j.ifacsc.2026.100384","url":null,"abstract":"<div><div>The behavior of a linear time-invariant system can be characterized entirely by measured input–output data that spans the vector space of all possible trajectories of the system relying on the fundamental lemma by Willems et al. However, useful <em>a priori</em> knowledge of the system is usually neglected. We propose a novel method for incorporating prior knowledge, specifically, known pole and zero locations, into a data-driven representation by constructing filters that pre-process the measured input–output data. To this end, a physics-informed data-driven predictor is introduced, where trajectories are obtained as linear combinations of the columns of a filtered block-Hankel matrix. We explicitly derive the output prediction error and show how leveraging prior knowledge reduces the impact of future noise realizations on output predictions and improves the accuracy of the initial state that is inferred from past data.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100384"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ensemble self-training deep partial least squares models for stable semi-supervised predictive learning and data analytics","authors":"Junhua Zheng ,&nbsp;Zhiqiang Ge ,&nbsp;Li Sun","doi":"10.1016/j.ifacsc.2026.100362","DOIUrl":"10.1016/j.ifacsc.2026.100362","url":null,"abstract":"<div><div>While deep learning has made significant achievements in the past years, it suffers from several serious shortcomings. Particularly, the performance of deep learning may be severely degraded under a small size of labeled training dataset, the case of which is quite common in industrial application scenarios although we are in the age of big data. In this paper, a semi-supervised deep model is proposed for predictive learning and data analytics, which is based upon the recently developed lightweight deep partial least squares model (PLS) structure. Precisely, the simple self-training strategy is used as the driving force to formulate the semi-supervised deep PLS model, which has no restriction in model structure and thus is flexible for predictive learning. In addition, to reduce the uncertainty of the self-training process, i.e. prediction error accumulation, different random seeds are introduced for model training, the results of which are combined together through an ensemble learning strategy. As a result, the predictive model becomes more stable and robust to those uncertainties introduced by both unlabeled data and the semi-supervised learning process. A real industrial example is provided for performance evaluation of the proposed method.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100362"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Online estimation of remaining time to recovery to enhance resilience using bond graph based power loss estimation","authors":"Mohd Faizan,&nbsp;Mahdi Boukerdja,&nbsp;Anne Lise Gehin,&nbsp;Belkacem Ould Bouamama,&nbsp;Sumit Sood","doi":"10.1016/j.ifacsc.2025.100357","DOIUrl":"10.1016/j.ifacsc.2025.100357","url":null,"abstract":"<div><div>Energy system resilience refers to the ability of systems to operate effectively during disruptive events. These disruptions occur when control mechanisms fail due to actuator saturation, triggered by faults or attacks with unpredictable behaviour. Maintaining system resilience relies on recovery control strategies. However, these strategies are often delayed, leading to severe system performance degradation. A novel indicator, Remaining Time to Recovery (RTTR), has been introduced in this work to address the delay in recovery control implementation. This indicator facilitates the implementation of the anticipatory recovery control strategies to address this delay. An innovative method for the online estimation of RTTR has been proposed, based on a hybrid approach that combines Bond Graph (BG) modelling and Machine Learning (ML). In the proposed work, the BG reference model interacts with system measurements and instantly estimates power losses caused by faults or attacks before the system’s performance is impacted. The ML layer, using linear regression (LR), processes the estimated power loss data to derive a prediction model of power loss evolution that is updated in real-time. RTTR is then predicted based on the initiation of power loss and the predicted evolution of that loss over time. The proposed methodology has been validated on a two-tank system using real-time Hardware-in-the-Loop (HIL) simulation with a Speedgoat target machine. The HIL simulations in different scenarios have been presented to demonstrate the reliability and accuracy of the proposed approach.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100357"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145885138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimal and robust control techniques for stability enhancement in a renewable integrated power system","authors":"Shuvo Dev ,&nbsp;Mehedi Hassan ,&nbsp;Naruttam Kumar Roy ,&nbsp;Rabiul Islam","doi":"10.1016/j.ifacsc.2025.100354","DOIUrl":"10.1016/j.ifacsc.2025.100354","url":null,"abstract":"<div><div>This study examines the design of a resilient control strategy for an IEEE 8-bus power system with renewable integration. It makes use of sophisticated control techniques such as Linear Quadratic Regulator (LQR), Linear Quadratic Gaussian (LQG), Sector-Bounded LQG (SBLQG), and Norm-Bounded LQG (NBLQG). By correcting model errors, the major goal of this study is to increase the power system’s resilience while preserving respectable performance indicators. To evaluate the efficacy of each control strategy, a thorough comparison is carried out using pole-zero plots, Bode plots, time-domain specifications, robust analysis, and statistical analysis. According to the pole-zero analysis, all control strategies have poles that are located in the left half-plane; the SBLQG and NBLQG strategies have the most leftward pole placements, which is a sign of better stability. The gain margin and phase margin consistently rise with each approach, according to Bode plot research, while the gain crossover and phase crossover frequencies also slightly increase. The controller’s enhanced robustness is evident in the 9.63% gain margin increases for LQG, 55.29% for SBLQG, and 86.79% for NBLQG when compared to LQR. In terms of time-domain performance, a decrease in rise time, peak time, and settling time is noted, while the percentage overshoot progressively diminishes in the sequence of LQR, LQG, SBLQG, and NBLQG. The percentage decrement in settling time for the controllers compared to LQR is 24.73% for LQG, 93.23% for SBLQG, and 98.06% for NBLQG, further highlighting their enhanced performance. The largest negative Cohen’s d values are observed in the comparison between LQR and NBLQG, with −24.4618 for GM and −18.9984 for PM, indicating a significant performance disparity. The results show that NBLQG is the most robust control strategy, exhibiting a modest settling time decrement. This research contributes to the field by illustrating how robust control methods, particularly NBLQG, effectively mitigate the impact of model uncertainties, thereby enhancing power system stability and performance in the presence of inaccuracies.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100354"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884546","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Structural identifiability in fractional-order networks","authors":"Alessandro Varalda ,&nbsp;Sérgio Pequito","doi":"10.1016/j.ifacsc.2026.100397","DOIUrl":"10.1016/j.ifacsc.2026.100397","url":null,"abstract":"<div><div>Identifying parameters of fractional-order discrete-time dynamical networks from input–output data is crucial for modeling complex systems in neuroscience and biology. Despite existing methods, fundamental conditions ensuring structural identifiability based solely on network structure remain unexplored. This paper establishes identifiability theory for fractional-order networks by proving that a network is structurally identifiable if and only if all its subsystems are identifiable. We introduce a graph-theoretical hierarchical algorithm that systematically partitions networks into identifiable subsystems via input–output reachability, enabling decomposition-based parameter learning with provable guarantees. Extensive simulation experiments validate that this approach preserves identifiability while achieving significant computational speedup, demonstrating scalability for large-scale systems.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100397"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147448753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Experimental Validation of the ACTIV Multi-Patient Mechanical Ventilation System","authors":"Lui Holder-Pearson ,&nbsp;J. Geoffrey Chase ,&nbsp;Yeong Shiong Chiew ,&nbsp;Geoffrey Shaw ,&nbsp;Bernard Lambermont ,&nbsp;Thomas Desaive","doi":"10.1016/j.ifacsc.2025.100351","DOIUrl":"10.1016/j.ifacsc.2025.100351","url":null,"abstract":"<div><div>Acute respiratory distress and respiratory disease often require patients be treated with mechanical ventilation (MV) and thus place extreme demand on intensive care units (ICUs). This burden can be unsustainably high in some periods, and particularly during pandemics, such as Covid-19. In low resource regions and countries, the result can be inequity, a problem addressable via simple technological innovation. Ventilator sharing over two or more patients has been proposed but strongly discouraged because it could not treat different patient needs and hindered individual patient monitoring. However, all these approaches ventilated patients in-parallel, each breathing at the same time.</div><div>A simple switching valve enables series breathing, one patient after the other. External, low-cost, and reusable sensor arrays enable individual monitoring, while low-cost adjustable pressure reducing valves allow pressure to be fully customised across two patients. This study uses an experimental test lung to experimentally demonstrate and validate the ability of such a system to balance ventilation across 2 simulated patients with very different lung compliances.</div><div>A method is presented to achieve equal tidal volumes in two lungs with differing compliances of 0.10 L cmH ₂O⁻¹ and 0.05 L cmH ₂O⁻¹. This goal requires driving and end-expiratory pressures of at least 20 cmH ₂O, which are clinically relatively high. The approach prioritises safety, ensuring more compliant lung is not over-ventilated during the process, reducing the risk of ventilator-induced lung injury (VILI). The system is compatible with different ventilators, and cost-effectively fabricated in low-resource settings. Strategies addressing key safety concerns, such as cross-contamination, sterilisation, and ventilator configuration, are also presented.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100351"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145697832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reference-model-based control including human torque estimation for cable-driven rehabilitation system","authors":"Nery U.O. Gutierrez ,&nbsp;T.M. Guerra ,&nbsp;Márcia L.C. Peixoto ,&nbsp;Paulo S.P. Pessim ,&nbsp;Antoine Dequidt ,&nbsp;Sebastien Delprat ,&nbsp;Vicenç Puig ,&nbsp;Sebastien Paganelli","doi":"10.1016/j.ifacsc.2026.100403","DOIUrl":"10.1016/j.ifacsc.2026.100403","url":null,"abstract":"<div><div>This paper presents a reference-model-based control strategy for human-interactive rehabilitation devices designed to ensure robust assistance during movement. The proposed approach combines feedforward and feedback actions to control the nonlinear system along physiotherapist-defined trajectories. The human torque, which represents the patient’s contribution to movement, is estimated in real-time using a Proportional-Integral Observer. This real-time estimation allows the system to adjust the level of assistance according to the user’s capabilities. Experimental validation in a prototype demonstrates the effectiveness of the proposed approach.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100403"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147384867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Determining switching strategies of multiprocesses via time partitioning and Pontryagin’s maximum principle","authors":"Nico Tauchnitz","doi":"10.1016/j.ifacsc.2026.100388","DOIUrl":"10.1016/j.ifacsc.2026.100388","url":null,"abstract":"<div><div>This paper addresses necessary optimality conditions in the form of the Pontryagin Maximum Principle for optimal multiprocesses. By a multiprocess, we understand a finite system of classical control processes, between which switching is freely permitted at any point in time. A particular challenge in studying multiprocesses is determining an optimal switching strategy. To this end, we present a new approach that identifies possible switching strategies with certain partitions of the finite time horizon. This enables us to incorporate the switching strategies as control variables and to interpret the problem of optimal multiprocesses as a classical control problem. The identification of an optimal switching strategy is thus achieved without a priori assumptions. However, the proposed approach imposes limitations in that the state trajectories must be continuous and no switching costs are allowed.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100388"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147384868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Iterative optimization in noisy dynamic environments via local pursuit","authors":"George Kartsiotis,&nbsp;Dimitrios Hristu-Varsakelis","doi":"10.1016/j.ifacsc.2026.100389","DOIUrl":"10.1016/j.ifacsc.2026.100389","url":null,"abstract":"<div><div>We discuss an iterative optimal control approach in which a multi-agent group of control systems finds locally optimal trajectories in noisy environments with time-varying terrain geometry and moving, deforming obstacles. Currently, the only method able to address the special noise-free case is the bio-inspired optimal control method termed ‘extended sampled local pursuit’. In that approach, agents follow one another, each one moving optimally towards its predecessor by periodically solving a fixed-final-time optimal control problem. We extend this method to settings where obstacles and agents are observed subject to noise, prove the convergence properties of our approach, discuss its computational complexity, and demonstrate its advantages in a series of simulations.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100389"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147384870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multisine input signal design for constrained, “plant-friendly” system identification of nonlinear systems","authors":"Sarasij Banerjee ,&nbsp;Eric Hekler ,&nbsp;Daniel E. Rivera","doi":"10.1016/j.ifacsc.2026.100371","DOIUrl":"10.1016/j.ifacsc.2026.100371","url":null,"abstract":"<div><div>This paper presents a methodology for optimizing “plant-friendly” multisine input signals to identify nonlinear dynamic systems under time-domain input and output constraints, without requiring a global parametric model <em>a priori</em>. The goal is to construct an informative dataset for open-loop, data-driven identification while selecting operational requirements. A weighted optimization framework is proposed to minimize the output crest factor resulting from a data-driven model, with penalties for violating input and output constraints. Model-on-Demand (MoD) estimation is employed to simulate outputs using prior data, effectively predicting nonlinear responses without global modeling. This MoD-based formulation enables evaluating output crest factors and output constraint compliance with modest modeling effort and improved impact. The resulting non-smooth, non-convex problem is solved using the Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm, which perturbs the multisine phase vector to achieve the desired performance efficiently. This method supports the concept of <em>identification test monitoring</em>, as illustrated in this paper. Within the identification test loops, each optimized excitation is applied to gather new estimation data, iteratively refining MoD-based output predictions and improving constraint satisfaction. The method’s effectiveness is demonstrated through a safety-critical case study on a Susceptible-Infected-Recovered (SIR) epidemiological network, showing that the optimized excitation yields highly informative data for identification while keeping the infection spread within safe limits.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"35 ","pages":"Article 100371"},"PeriodicalIF":1.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146077766","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}