ISA transactions最新文献

This paper investigates the practical bipartite tracking consensus for uncertain nonlinear high-order multi-agent systems with input saturation and time-varying input delay over signed switching topologies. To address input saturation and time-varying input delay simultaneously, an auxiliary system is developed. The states and the total disturbance of the considered systems are estimated by designing a dynamic event-triggered extended state observer (ESO). Under the framework of the command filtered backstepping, a dynamic event-triggered controller is proposed utilizing the states of the ESO. Then the practical bipartite tracking consensus can be guaranteed under the proposed controller. Meanwhile, the Zeno behavior is excluded. Finally, the validity of the obtained results is shown by a numerical example.

This paper develops a preset-trajectory-based adaptive output-feedback strategy for decentralized prescribed-time tracking of uncertain strict-feedback nonlinear systems subject to interconnections and dead-zone inputs. The framework assumes that unmatched nonlinear functions, nonlinear interconnection terms, and the parameters of dead-zone input nonlinearities are completely unknown. The primary contributions include designing local tracking-error trajectories (i.e., local preset trajectories) using local output-feedback signals and developing a novel design strategy with a time-varying function to initialize parameter estimation errors to zero without requiring zero initial estimates. Neural-network-based state filters reconstruct the unmeasured states, while an adaptive dead-zone inverse approximation compensates for the unknown dead-zone nonlinearity. A decentralized output-feedback controller is designed to achieve practical prescribed-time stability, with transient behavior shaped by the constructed preset trajectories. The proposed design explicitly avoids the singularity that may arise in the adaptive dead-zone inverse approximation due to parameter estimates approaching zero. This work rigorously analyzes the boundedness of the closed-loop signals and practical prescribed-time stability of the local tracking errors, based on the zero initial condition of the Lyapunov function. The simulation results comparing the proposed approach with existing methods demonstrate its effectiveness and advantages.

In this paper, a novel event-triggered-based decay aggregation efficient model predictive control (DAEMPC) problem is investigated for nonlinear systems represented by interval type-2 (IT2) T-S fuzzy models subject to cyber attacks and actuator saturation. First, to make full use of communication resources, an adaptive event triggered (AET) strategy is applied to determine the data transmission in the sensor to controller link. A Bernoulli random process is introduced to denote the denial-of-service (DoS) attack, and the polytopic description method is utilized to characterize the actuator saturation. Second, the efficient model predictive controller concerning the decay aggregation approach is designed for the considered nonlinear networked control system (NCS). It involves offline solving feedback control law and designing ellipse feasible sets whose projections are vertical in the x-space, and online optimizing the perturbation variable instead of the whole performance objective function. Different from the previous studies, the presented AET-based DAEMPC algorithm not only compensates for the deficiencies in the communication network, but also enlarges the initial feasible set and reduces the computational burden. Finally, the validity of the presented algorithm is illustrated through the simulation of continuous stirred tank reactor (CSTR).

This study mainly focuses on an optimized leader-follower formation problem for a group of nonholonomic wheeled mobile vehicles (NWMVs). By systematically integrating the critic-actor reinforcement learning (RL) and adaptive neural network (NN), a distributed cooperative formation scheme for the multi-nonholonomic wheeled mobile vehicles (MNWMVs) consisting of a kinematic controller and a dynamic torque controller is proposed. The Hamilton-Jacobi-Bellman (HJB) equation, regarding the performance index function, possesses highly nonlinear and strongly coupled characteristics. It is challenging to solve the HJB equation to acquire the optimized formation protocol of multiple NWMVs, as it possesses under-actuated and nonholonomic Lagrange dynamic properties. Significantly, the key feature of the developed optimized formation tracking algorithm for MNWMVs is an adaptive identifier integrated into the critic-actor RL strategy. It effectively addresses the uncertainties associated with Lagrange dynamics. Furthermore, the developed optimized formation scheme is greatly simplified due to the RL training laws obtained from the negative gradient of a simple positive function. Finally, numerical simulations and physical experiments are performed to validate and demonstrate the theoretical results.

The blast furnace ironmaking process exhibits periodic behavior, time-varying delays, and complex spatiotemporal coupling, making it difficult to achieve real-time monitoring of gas flow distribution. In response to these challenges, this paper proposes a soft sensor-driven proximal policy optimization (PPO) framework with spatiotemporal periodic modeling and dynamic memory (SPDM-PPO) for synergistic predictive control. Firstly, to overcome the modeling inaccuracies caused by dynamic coupling and uncertain time delays, a dynamic time-delay optimization method is developed by embedding spatial regularization into mutual information, eliminating the hysteresis effects. Subsequently, a dual-encoding Transformer network is designed, which incorporates both absolute and periodic positional encodings to capture spatiotemporal periodic patterns and global dynamics. Then, considering the issues of information redundancy and memory obsolescence in periodic state representation, a dynamic periodic state memory (DCSM) mechanism is proposed by aggregating dual-threshold memory optimization and attention-weighted. Furthermore, to achieve dynamic closed-loop predictive control of gas flow distribution, a cooperative dual-optimizer-trained PPO strategy and the DCSM are embedded, along with a long short-term memory (LSTM) encoder-decoder. Finally, extensive experiments conducted on real-world BF industrial data robustly validate the effectiveness and superiority of the proposed framework.

In this paper, a disturbance observer-based adaptive event-triggered model predictive control (DAEMPC) method is proposed for a class of nonlinear systems with constraints and bounded disturbances. First, a disturbance observer is employed to actively compensate for disturbances. Leveraging the space decomposition technique, the disturbances are divided into the matched parts and the remaining unmatched parts. The matched disturbances are compensated using the pre-designed disturbance observer. To address the effects caused by the remaining unmatched disturbances, a bounded controller and an optimal controller with an adaptive event-triggered mechanism are respectively designed based on whether the system state resides within the stable region. The larger terminal stability estimation set is calculated based on the bounded controller. Furthermore, rigorous theoretical analysis is performed to prevent Zeno behavior. Finally, the simulation results for two numerical examples verify the effectiveness of the proposed algorithm.

In this paper, a data-driven distributed alternating optimization approach to optimal fault detection is proposed for dynamic processes based on canonical correlation analysis (CCA). The focus of this method is to reduce the uncertainties caused by measurement noise using relevant information from the neighboring subsystems. Specifically, the average consensus algorithm is used in the alternating optimization algorithm to calculate the CCA parameters, thereby enabling each subsystem to update the parameters simultaneously. Then, a distributed residual generator can be constructed using the obtained CCA parameters for the fault detection purposes. Compared with the centralized methods, the communication cost between nodes is reduced and the computation efficiency is improved by the proposed distributed approach. Based on it, case studies on the hot rolling mill process and Tennessee Eastman process are used to demonstrate the proposed method.

Robust online composition estimation is crucial for sustaining energy efficiency in industrial distillation processes, especially under dynamically drifting and unobserved conditions. Classical model-based estimators often encounter limitations in adapting to nonlinearities, signal drift, and measurement noise, while pure data-driven techniques fail to generalize to unseen trends and process drifts, resulting in suboptimal servo and regulatory performances. To overcome these limitations, this study proposes a hybrid data-driven soft sensor framework that integrates a Quadratic Program-based Constrained Kalman Estimator (QP-CKE) with Piecewise Linear Regression (PLR), along with an Adaptive Window (AW) extension. The AW-PLR dynamically adjusts regression window length based on a Cumulative Sum (CUSUM) derived nonlinearity index, enabling better sensitivity to transients and non-stationary process behaviour. The proposed soft sensors are validated for an ethanol-water mixture separation in a pilot-scale laboratory distillation column. Performance is benchmarked against Extended Kalman Estimator (EKE), static QP-CKE, and a Support Vector Machine (SVM)-based regression model. Quantitative evaluation shows that the proposed AW-PLR based QP-CKE achieves approximately 35 % lower RMSE, 22 % higher SNR, and 25 % faster computation time, while also offering noise resilience and process interpretability compared to its counterparts. The proposed hybrid soft sensors demonstrate enhanced adaptability, computational efficiency, and robustness, supporting their suitability for integration into real-time soft sensing and control frameworks.

Multirate sampled data (MRSD) dynamic systems are abundant in modern engineering systems. The inconsistent sampling rates cause data asynchrony and alter system properties, negatively impacting fault diagnosis with delayed and missed detections. In this paper, a fast rate fault detection scheme for dynamic systems is proposed, which is directly driven by MRSD. Firstly, the lifting technique is employed to transform asynchronous MRSD into single but slow rate sampled data. An auxiliary lifted output is constructed to compute a parity vector via subspace identification, facilitating a multi-dimensional diagnostic observer satisfying the Luenberger conditions. Then a post filter addresses causality constraint, allowing fast rate residual generation. Further, a fast rate residual evaluation scheme is developed. The effectiveness and superiority of the proposed scheme are demonstrated by a heating, ventilation and air conditioning (HVAC) example.

This paper examines the prescribed-time consensus of multi-agent systems under time-varying communication delays and Denial-of-Service (DoS) attacks. Considering a general class of DoS attacks with limited duration, a novel control protocol accompanied by time-varying node delays communication delays and integral action is proposed to guarantee the secure prescribed-time consensus. Moreover, we propose a controller to achieve prescribed-time consensus by utilizing the Artstein's reducing transformation, effectively addressing the challenges posed by time-varying delays. By using the comparison principle and Lyapunov stability theory, consensus convergence properties are analyzed, and sufficient criteria are obtained. Furthermore, a distributed prescribed-time observer is introduced to guarantee that all follower agents obtain the leader's state information within the prescribed time, even if only a subset initially has access. To conclude, a numerical simulation is offered to substantiate the robustness and implementation of our theoretical insights.