To handle the nonlinear consensus problem, a distributed model predictive control (DMPC) scheme is developed via parametric sensitivity. A two-stage input computation strategy is adopted for enhancing optimization efficiency. In the background stage, each agent first establishes its next-step optimization problem based on communication topology, and then performs distributed optimization to calculate the future inputs. In the online stage, all the agents build their sensitivity equations based on new information. Three variants of sensitivity equation are developed based on the level of communication load capacity, and the corresponding computation strategies are proposed. After solution, the background inputs are corrected and implemented. The optimality and robustness of the proposed algorithm are rigorously derived. Finally, the superiority of this DMPC scheme is demonstrated in the multi-vehicle system with two different topologies.

A degradation modeling method of marine gas turbines for the overhaul cycle is proposed to address the problem of insufficient data sets for fault diagnosis and trend prediction algorithm validation. First, a nonlinear model of the marine three-shaft gas turbine gas path was established. The degradation path and component degradation models were subsequently obtained. The distribution of the washing cycle and degradation factors in the overhaul cycle were solved using an optimization algorithm, and degradation data in the washing cycle were obtained. The model's feasibility is validated with a segment of actual degradation data. The change rule of the gas turbine operating data during the overhaul cycle was also obtained. The degradation data of marine gas turbines under different boundary conditions are simulated using the established degradation model. This model provides data sets essential for validating fault diagnosis and trend prediction algorithms. Furthermore, it provides a reference for modeling the degradation of other mechanical equipment.

Efficient and accurate real-time estimation of power system synchronization is quite important for its safety control and operation. However, signal sensing failure, electromagnetic interference, system delay, etc., will cause the sensor gain degradation. To furnish a dependable method for dynamic estimation in power grid synchronization amid sensor gain degradation, this research presents a robust estimation system capable of monitoring and tracking the frequency, voltage phase angles, and magnitudes. Firstly, the random degradation of measurement data is characterized by a discrete distribution within the range [0,1]. Secondly, the state space model of sensor gain degradation is established. Subsequently, a novel modified fault-tolerant extended Kalman filter (MFTEKF) is developed under the recursive estimator framework. Finally, extensive experimental results definitively demonstrate that the proposed MFTEKF can accurately monitor the dynamic characteristics of the power grid.

The topside separation system plays a pivotal role in the treatment of produced water within offshore oil and gas production operations. Due to high-humidity and salt-infested marine environments, topside separation systems are susceptible to dynamic model variations and valve faults. In this work, fault-tolerant control (FTC) of topside separation systems subject to structural uncertainties and slugging disturbances is studied. The system is configured as a cascade structure, comprising a water level control subsystem and a pressure-drop-ratio (PDR) control subsystem. A fault-tolerant H control framework is developed to cope with actuator faults and slugging disturbances. To enhance control performance in the presence of actuator faults and model uncertainties while reducing sensitivity to slugging disturbances, the fault-tolerant H control problem for the topside separation system is established as the two-player differential game problem. In addition, a Nash equilibrium solution for the fault-tolerant H control problem is achieved through the solution of the game algebraic Riccati equation (GARE). A model-free approach is presented to implement the proposed fault-tolerant H control method using off-policy reinforcement learning (RL). Simulation studies demonstrate the effectiveness of the solution.

This paper focuses on sequential fusion estimation for multi-rate multi-sensor nonlinear dynamic systems with heavy-tailed noise and missing measurements. On the basis of Bayesian inference, a sequential Student's t-based unscented Kalman filter (SSTUKF), together with its square-root form (SR-SSTUKF), is proposed by using the unscented transform to calculate Student's t weighted integrals. Considering the nonstationary measurement noise and/or accumulated computation error, adaptive factors are introduced by the t-test to suppress uncertainties. Additionally, the complexity computation and convergence analysis of the SR-SSTUKF are presented. The validity and robustness of the proposed sequential fusion method are illustrated by an example of agile target tracking. Simulation results indicate that the SR-SSTUKF with adaptive factors can further enhance accuracy and yield reliable estimations.

This paper presents an active disturbance rejection control (ADRC) approach for three-phase four-legs voltage source inverters (FL-VSIs) in a standalone renewable energy resources (RES)-based islanded microgrid. The key purpose of the proposed approach is to improve the control robustness against load-side disturbances, power supply parameters uncertainties, and faulty operating conditions. Indeed, a notable benefit of ADRC is its ability to operate effectively without the need for precise knowledge of disturbance characteristics or accurate modeling and FL-VSI parameters. As compared with conventional PI controllers, this advanced control strategy allows improving the voltage waveforms quality and conforming to existing power quality standards and metrics while using only output voltage sensors. Extensive simulations on Matlab/Simulink software have been conducted to assess the effectiveness of the proposed approach.

To increase the adaptability of bridge construction equipment in high-altitude settings, this study examines a magnetorheological (MR) damper designed for cable-stayed climbing robots. Initially, a novel damper incorporating a spring-MR fluid combination and three magnetic circuit units is developed. A robot-cable-wind coupling dynamic model is subsequently formulated via Hamilton's principle, based on force analysis. The simulation results indicate that the damper's maximum output force is 204.60 N, with optimal working currents of 0.2 A (Force 4) and 0.4 A (Force 7). To verify the analysis, testing is conducted using an MR damper. The results reveal an average relative error of 4.60% for the actual output damping force. When mounted on the robot, the climbing speed range, average relative error, and maximum relative error are controlled within 0.66 mm/s, 0.78% and 2.5%, respectively. This approach allows for the rapid selection of suitable working currents and markedly enhances the climbing stability of the robot.

The stability of the suspension is a key challenge for the application and promotion of electromagnetic suspension technology, especially when it operates in conjunction with a flexible structure, which significantly increases the system's complexity. This paper abstracts the characteristics of the coupling conditions between an electromagnetic suspension system and a flexible structure and designs and constructs an experimental apparatus that includes an electromagnet and a simulated flexible structure with adjustable stiffness and inertia. Based on the Lyapunov method, the central manifold theorem, and the Poincaré method, the stability of the electromagnetic suspension system and the conditions for Hopf bifurcations are derived. Finally, through reasonable experimental design and data analysis, the correctness of the theoretical analysis conclusions is validated, providing references for the engineering applications of electromagnetic suspension systems.

This paper considers the exact controllability of Itô stochastic systems with input delay. In particular, one delay-free controller and one delayed controller are involved in the systems which complicates the study due to the inconsistency of adaptiveness caused by input delay. The main contribution of this paper is to provide the necessary and sufficient Gramian matrix condition and the necessary Rank condition for the exact controllability. The key is to solve the backward stochastic differential equations (BSDEs) with input delay.

In response to the issues of low tracking accuracy and poor robustness in the trajectory tracking control of a permanent magnet spherical motor (PMSpM), an active disturbance rejection control (ADRC) scheme combining neural networks is put forward in this research. The unknown total disturbance is approximated by employing a radial basis function (RBF) neural network, with weights updated by an adaptive law and compensated for through the nonlinear feedback loop. This approach addresses the problem of performance degradation of the extended state observer under severe total disturbance, thereby ensuring accurate tracking of the PMSpM. Comparative simulations are accomplished to evaluate the performance of the RBF-ADRC scheme in enhancing disturbance rejection capability and robustness. Experimental results from the planar circular motion experiment on the PMSpM test platform validate the application value of the scheme.

In order to effectively model data online, a learning model must not only have the high adaptability of dynamic data but also keep the low complexity to meet the online computing requirements. In this paper, a novel multi-innovation online sequential extreme learning machine (MIOSELM) and its kernel version called multi-innovation kernel online sequential extreme learning machine (MIKOSELM) are proposed to establish the online estimation models based on p latest samples using the multi-innovation theory. Besides, a modified whale optimization algorithm (MWOA) is introduced to optimize the execution parameters of our algorithms and is capable of automatically searching a proper p as the practical need, which can further improve the adaptability performance of the online learning models. Finally, two different datasets (the UCI dataset and KDD99 dataset) are used to evaluate the superiority of our methods. Experimental results show that the accuracy, F-score, and G-mean of MIKOSELM are 98.25%, 98.11% and 98.63% on WDBC from the UCI dataset, and are 83.61%, 75.96% and 70.97% on the KDD99 dataset respectively. Besides, our MIKOSELM based on MWOA achieves F-score of 94.28% and 76.73% on Musk from the UCI dataset and the KDD99 dataset. These results validate the effectiveness of our proposed methods.

The article proposes a novel state-feedback control method for a multiple-input multiple-output (MIMO) nonlinear system with actuator faults and input quantization. The innovation of the design approach lies in the utilization of fuzzy logic systems (FLSs) to approximate the uncertain intermediate virtual control laws, thereby achieving a simplified virtual control design form. Additionally, finite-time control is employed to enhance the system's response speed. Different from the existing literatures, the adaptive control scheme of partial loss fault gain is integrated with input quantization, which completes the unknown gain estimation and avoids the assumption condition of unknown control gain. The theoretical analysis combined with Lyapunov stability analysis shows that the tracking error can converge regardless of whether the system experiences a fault, while the closed-loop signal remains stably bounded for a finite time. Finally, the simulation results of the quadrotor unmanned aerial vehicle (UAV) attitude system indicate that this control scheme is effective.

A fully actuated system (FAS) approach integrated with adaptive robust integral of the sign of the error (ARISE) feedback control strategy is proposed for multi-input multi-output nonlinear systems in the presence of both external disturbances and parametric uncertainties. Owing to an inability to eliminate unmeasured disturbances and model inaccuracies simultaneously, the existing results based on the FAS approaches are typically limited to the uniformly ultimate boundedness of the tracking errors. To achieve the asymptotic tracking performance confronted with parametric uncertainties and time-varying disturbances, an ARISE feedback controller with desired compensation is synthesized to suppress the adverse effects arising from nonlinearity and uncertainty of the system. The improvements compared to the traditional RISE feedback control are attributed to two aspects: (i) the feedback gains in the RISE term are adjustable-online without having to know the prior bounds of disturbances and their time derivatives; (ii) a desired compensation-based adaptive feedforward term, primarily employing the desired trajectories in place of the measured states, could weaken the underlying interaction between the adaptive compensation and robustness part. A rigorous stability analysis is provided to demonstrate that the system state can asymptotically track a bounded desired trajectory in spite of bounded disturbances and parametric uncertainties. Comparative simulations on an under-actuated planar manipulator, possessing an equivalent multi-order FAS model, have been conducted to verify the effectiveness and merits of the developed controller. Experimental validation on a two-wheeled self-balancing robot is also provided to show the feasibility of the proposed approach.

The coupled relationship between inputs and outputs in multiple-input multiple-output (MIMO) systems, as well as the multiplicative uncertainties caused by multiplicative faults, increases the complexity of fault diagnosis (FD) and fault-tolerant control (FTC). Research has indicated that coprime factor uncertainties are suitable for modeling multiplicative uncertainties. This paper presents an FD and FTC strategy for MIMO systems based on the ν-gap metric technique within the coprime factorization framework. In the offline phase, the ν-gap metric-based hierarchical clustering method is designed to classify fault samples. Next, core systems and boundary systems are calculated for each fault category, and corresponding residual compensation controllers are designed. In the online phase, by computing the relevant ν-gap metric values, the fault severity of the real-time system is determined, and the core system with similar dynamic behaviors is identified. This FD result drives the switching of residual compensation controller, achieving FTC and ensuring system stability and robustness. This strategy eliminates the need for online solving of fault-tolerant controller, saving computational resources. Finally, the ν-gap metric-based FD and FTC strategy is validated with simulations on a three-phase voltage source inverter system.

Hydraulic load-sensitive systems (HLSS) are widely used for high power density and energy efficiency. This study introduces an adaptive, energy-efficient HLSS with a valve-controlled variable motor. The system faces challenges from non-linearities, including internal higher-order dynamics due to displacement changes and external unknown disturbances, which hinder precision applications. To address this issue, this study explores HLSS principles to develop an accurate system model. Subsequently, an adaptive robust motion control that considers displacement compensation (DCARC) is proposed using the established model. DCARC can learn unknown parameters online and compensate the model more accurately to improve control accuracy. Experiments show that considering the higher order dynamic effects caused by displacement in the system can improve model accuracy and effectively reduce the burden of parameter adaptation and robust feedback terms. High-precision and energy-efficient HLSS motion is verified and realized in the study. The control accuracy of DCARC is 19.4% higher than that of conventional adaptive robust control (ARC). Under experimental conditions, the proposed system can improve energy efficiency by up to five times compared to valve-controlled fixed displacement motor systems (VFDS).

This paper presents a map-free navigation approach for industrial automatic mobile robots (AMRs), designed to ensure computational efficiency, cost-effectiveness, and adaptability. Utilizing deep reinforcement learning (DRL), the system enables real-time decision-making without fixed markers or frequent map updates. The central contribution is the Heuristic Dense Reward Shaping (HDRS), inspired by potential field methods, which integrates domain knowledge to improve learning efficiency and minimize suboptimal actions. To address the simulation-to-reality gap, data augmentation with controlled sensor noise is applied during training, ensuring robustness and generalization for real-world deployment without fine-tuning. Training results underscore HDRS's superior convergence speed, training stability, and policy learning efficiency compared to baselines. Simulation and real-world evaluations establish HDRS-DRL as a competitive alternative, outperforming traditional approaches, and offering practical applicability in industrial settings.

This article presents a cascade controller for the quadrotor to track the desired trajectory effectively. Unlike previous approaches, this method avoids simplification and linearization assumptions, making it applicable in a wider range of scenarios. A novel linear quadratic tracking method is utilized, which takes into account both process noise and measurement noise while maintaining a model-free nature. Furthermore, the stability analysis of this stochastic method is thoroughly investigated. In terms of attitude control, a model-free approach is adopted. The Deep Deterministic Policy Gradient (DDPG) algorithm is implemented, leveraging an actor-critic network to handle the nonlinearities associated with attitude control. This model-free approach eliminates the need for an accurate model of the quadrotor's dynamics. Simulations are conducted to evaluate the performance of the proposed controller, and the results demonstrate its ability to effectively control the quadrotor, ensuring accurate trajectory tracking and stability.

Accurate prediction of remaining useful life (RUL) is crucial for predictive maintenance of industrial systems. Although data-driven RUL prediction methods have received considerable attention, they typically require massive run-to-failure (R2F) data which is often unavailable in practice. If not properly addressed, training with a limited number of R2F trajectories not only leads to large errors in RUL prediction, but also causes difficulty in quantifying the prediction uncertainty. To address the above challenge, this paper proposes a Bayesian ensemble RUL prediction method that combines mode-dependent relevance vector machine (RVM) and trajectory similarity. Firstly, the proposed approach clusters historical R2F trajectories of unequal lengths into different degradation modes, and constructs RVM and similarity based predictions with improved accuracy by using mode-dependent libraries of kernel functions and similar trajectories. Secondly, the proposed Bayesian ensemble scheme fuses the RVM and similarity based predictions, and quantifies the associated prediction uncertainty even though the number of historical R2F trajectories are limited. In two case studies involving bearings and batteries, using only 11 and 16 R2F trajectories as training data, respectively, the proposed method reduces the mean absolute percentage error of RUL prediction by more than 20% compared to three existing methods.

The performance of kernel adaptive filtering algorithms (KAFs) with nonlinear recurrent structures surpasses traditional KAFs, attributed to the nonlinear contribution of feedback. Nevertheless, the existing nonlinear recurrent KAFs primarily focus on a single feedback output, potentially limiting their latent filtering capabilities. In this paper, we introduce a novel alternative, named nonlinear recurrent kernel normalized least-mean-square with multiple feedbacks (NR-KNLMS-MF), which leverages the information from multiple feedback outputs. Additionally, to tackle the computational complexity challenges associated with KAFs, we integrate random Fourier features (RFF) into NR-KNLMS-MF, resulting in an efficient variant called as RFF-NR-KNLMS-MF. Furthermore, we conduct a theoretical analysis of the mean-square convergence for RFF-NR-KNLMS-MF. Simulation results on time-series predictions demonstrate the superiority of our proposed algorithms over other competing alternatives, validating their effectiveness.

This work addresses the discrete-time containment control problem for first-order and second-order systems, incorporating position constraints, nonuniform time delays, and switching topologies. Projection operators are introduced to maintain agent states within the position constraints. Then, model transformation methods, along with stochastic properties of matrices, are employed to handle coupled nonlinearities stemming from position constraints and time delays. By using local information, distributed projection-based control schemes are proposed for multi-agent systems. Theoretical analysis shows that all followers ultimately converge into the convex hull spanned by leaders if the union of the communication topologies contains a directed spanning tree within each specified time interval. Finally, the theoretical results are verified by numerical simulations.

Inertial reference unit (IRU) has been recognized as an effective method for photoelectric tracking and pointing system to suppress the disturbance. Its flexible support usually results in a low-frequency mechanical resonance within the bandwidth. Experimental identification indicates an obvious resonant peak at near 36.7 Hz with an amplitude of 26.6 dB. In this paper, disturbance observer based on band-pass filter (BPF-DOB) is combined with a notch filter to suppress the mechanical resonance. The asymmetry of the resonance peak and variation of the controlled object over the time are both addressed. Simulation and experimental results show that the stability accuracy of IRU system is improved to be 1.09 μrad, indicating an improvement of 9.9 % relative to the traditional method and high robustness for time-varying resonance.