ISA transactions最新文献

This article investigates the prescribed performance angular position tracking control of a rehabilitation exoskeleton with uncertain dynamics, unknown disturbances, actuator faults, and flexible joints. First, a novel appointed-time prescribed performance function (APPF) is proposed, wherein the convergence time can be accurately pre-specified. Then, the APPF is integrated with an improved sliding mode reaching law to develop an appointed-time fault-tolerant prescribed performance control (AFPPC) strategy. The attractive features of the proposed AFPPC include: 1) guaranteeing the appointed-time convergence with faster transient response and higher steady-state performance, despite the presence of uncertain dynamics, unknown disturbances, and actuator faults; 2) enhancing the robustness of the system and suppressing the vibration of the flexible joints using an improved reaching-law-based sliding mode prescribed performance framework. These features are capable of improving the safety of the rehabilitation therapy. Lyapunov-based theoretical analysis and comparative simulations demonstrate the benefits of the proposed method.

In this paper, we propose a predefined-time nonsingular sliding mode control (SMC) method based on emotional neural networks (ENN) for disturbed nonlinear systems. Firstly, based on the predefined-time stability (PDTS) theory, a new sufficient condition is constructed. By introducing a novel adaptive continuously adjustable exponential term into the Lyapunov stability lemma, this modification abandons the traditional fixed parameterization of exponential terms. Secondly, by introducing ENN to approximate the equivalent control term, a nonsingular SMC is designed, meanwhile solving the issue of unknown dynamics. Using Lyapunov theory, this paper rigorously proves the system's PDTS in both the reaching and sliding phases. Finally, simulations on the inverted pendulum and deep-sea vehicle manipulator show that they can track the target trajectory in 0.22 and 0.23 s respectively, outperforming comparison methods.

This paper presents a double-risk-constrained linear quadratic regulator (DRC-LQR) for practical control of partially observable systems under non-Gaussian and biased disturbances. Unlike conventional risk-neutral or single-risk formulations, the proposed approach jointly constrains state and output variability through a computationally tractable primal-dual optimization framework, enabling explicit compensation for noise skewness and heavy tails. As the first infinite-horizon formulation ensuring joint stability and constraint satisfaction under general non-Gaussian conditions, the DRC-LQR achieves both theoretical rigor and real-world feasibility. Comprehensive simulations on aircraft flight control and voltage regulation tasks demonstrate over 60% improvement in regulation accuracy, 93.7% faster convergence, and 99.8% constraint satisfaction compared with standard LQR, confirming its superior robustness and practicality. These results establish DRC-LQR as a systematic and implementable extension of LQR, advancing risk-sensitive control design for safety-critical systems subject to extreme disturbances.

This paper is concerned with the protocol-based state estimation issue for nonlinear time-delayed multiplex networks (NTD-MNs) suffering from deception attacks. The classical Round-Robin (RR) protocol stands out for its advantage in scheduling the data transmission of complex networks (CNs) with limited communication bandwidth. Nevertheless, the excessive sacrifice of the observability of the CNs leads to the inevitable degradation of estimator performance. A novel scheduling mechanism called the Grouped-Round-Robin (GRR) protocol is thus first proposed to govern the data transmission of nodes within the NTD-MNs, and simultaneously overcome the deficiency of the conventional RR protocol. Besides, considering the openness of the communication environment, cyber-attacks are usually inevitable, and stochastic deception attacks modeled with a Bernoulli process are thus taken into account. Employing the proposed GRR protocol, a state estimator is established which considers the impact of deception attacks on the measurement output. Then, sufficient conditions are derived to ensure that the estimation error for the discussed system is ultimately bounded in mean square. Finally, theoretical results are corroborated through numerical simulations.

To improve the tracking control performance in task space of the Stewart parallel mechanism (SPM) with disturbances and uncertainties, a fixed-time sliding mode composite control with prescribed performance (FxT-SMCC-PP) method is developed. First, by constructing a sliding mode term adaptive to the changes of the lumped uncertainty, and embedding fixed-time performance parameters, a fixed-time adaptive sliding mode disturbance observer (FxTASMDO) is designed. Second, the SPM's tracking error is constrained by an asymmetric hyperbolic cosecant prescribed performance function and transformed into a steady equivalent error by modifying the constraint transformation function. Based on the equivalent error, a fixed-time sliding mode control with prescribed performance is designed and combined with FxTASMDO to form the FxT-SMCC-PP, which can achieve the SPM's fast and non-overshoot transient response, high steady-state tracking accuracy, and reduction in control energy consumption. Finally, the effectiveness of FxT-SMCC-PP is validated through simulation and prototype experiments.

This article focuses on designing an innovative security control methodology for linear networked control systems under deception attacks, which are stochastically attacked by two unknown-bounded deception signals. Firstly, an ambiguity set is introduced to characterize all potential deception signals that satisfy the identical mean-covariance constraints instead of bounded constraints. Then, chance constraints concerning system state and control variables are formulated to mitigate conservatism in the security control design. On the basis of the principle of distributionally robust optimization, chance constraints are addressed by handling a deterministic convex reformation problem. Subsequently, a stochastic model predictive control approach is deployed to realize the recursive feasibility and convergence of the controlled model. Finally, different scenarios of malicious attacks concerning the DC-DC boost converter are presented with the aim of validating the superiority of the designed approach.

Taking the H₂ norm of the tracking error signal as the performance index, this study examines the output tracking performance limits of networked systems with bandwidth-limited communication channels. Two linear time-invariant filters are used to characterize the bandwidth constraints and additive colored Gaussian noise (ACGN), respectively. By invoking some factorization techniques and Youla parameterization of two-parameter controllers, the fundamental tracking performance limitations are achieved. From the exact expressions, it is evident that there exists a close relationship between performance limitations and the essential features of the plant. Further, it is also concluded that the limited bandwidth and ACGN generally have adverse effects on tracking performance. Finally, the theoretical results are verified through a practical plant example. For a linear dual-freedom vehicle system, the results also show the effectiveness of the proposed method.

In this article, a sampled-data event-triggered resilient consensus tracking control scheme is proposed for multiple flexible manipulators with actuator failures. First, for each flexible manipulator, the finite-time observer is designed to achieve state estimation of the virtual leader. Then, the periodic disturbances are effectively estimated by an iterative learning method. Furthermore, a resilient consensus control scheme is developed by integrating a sampled-data event-triggered mechanism, and it have been proved from the obtained stability results that the proposed control scheme can achieve consensus control and vibration suppression of multiple flexible manipulators while reducing communication consumptions. Finally, the effectiveness of the proposed control method is demonstrated through simulation results.

This paper addresses the load frequency control problem subjected to communication delays, actuator non-linearity, and parameter uncertainties using a Model Assisted Reduced-order ADRC (MRADRC). The approach makes use of minimal plant knowledge in the Extended State Observer (ESO) design with reduced-order so that the observer becomes delay aware, thereby improving its estimation accuracy. A two-stage tuning approach is introduced to tune the parameters of the controller. In the first stage, the Walrus multi-objective optimizer obtains a set of Pareto-optimal solutions found by minimizing frequency deviation metric (IAE) and control signal variation (TV) subjected to mixed robustness level 2≤ϵ≤5. In the second stage, multi-criteria decision-making based ranking methodology is used to obtain final optimal controller and observer bandwidths. The proposed method is implemented on two numerical studies. Study 1 focuses on the implementation of the proposed method on single-area LFC (non-reheat, reheat, hydro) plants under four different scenarios involving nominal, perturbed, and non-linearity cases. In Study 2, a more reliable benchmark system, IEEE 39-bus New England system, is considered with the controller tested for cases involving random load and variable delays. In both studies, it was observed that MRADRC exhibits considerable improvements in reducing frequency deviations and its peak level compared with PID/FOPID/PI/H_∞ methods, while maintaining robustness level (ϵ) at the desired level and achieving a good delay margin.

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.