International Journal of Adaptive Control and Signal Processing最新文献

This study focuses on addressing the challenge of adaptive finite-time control for nonstrict-feedback nonlinear systems subject to input delay and saturation. Neural networks (NNs) are utilized to handle unknown nonlinear functions, and Padé approximation is employed to effectively manage input delay. To mitigate the issue of “explosion of complexity,” the command filter method is applied. By leveraging command filter technology and backstepping technique, an adaptive finite-time control scheme is developed using NN approximation. The proposed control scheme demonstrates that the closed-loop signals achieve semi-global practical finite-time stable (SGPFS), ensuring that the tracking error converges within a finite time to a small region around the origin. The effectiveness of the proposed scheme is validated through two simulation examples.

The Kalman filter is an important technique for system state estimation. It requires the exact knowledge of system noise statistics to achieve optimal state estimation. However, in practice, this knowledge is often unknown or inaccurate due to uncertainties and disturbances involved in the dynamic environment, leading to degraded or even divergent Kalman filtering solutions. This paper proposes a novel method of H∞ filtering-based on adaptive random weighted estimation to address this issue. It combines the H∞ filter with random weighted concept to estimate system noise statistics. Random weighting theories are established based on the state estimate and state error covariance of the H∞ filter to estimate both process noise statistics and measurement noise statistics. Subsequently, the estimated system noise statistics are fed back into the Kalman filtering process for system state estimation. Simulation and experimental results show that the proposed method can effectively estimate system noise statistics, leading to improved accuracy for system state estimation.

Automated categorization of electrocardiogram (ECG) waveforms using deep learning (DL) methods has garnered considerable attention in recent research. However, prevalent DL networks encounter challenges including overfitting, class imbalance, limitations in deeper network training, and high computational demands. To address these issues, this study proposes an Automated ECG Arrhythmia Classification framework employing the Reinforced Visual Geometry Group-27 (REF-VGG-27). Initially, the framework encompasses preprocessing steps such as denoising, R-peak identification, data balancing, and cross-validation. For automatic feature extraction and classification, two DL architectures are suggested: a novel hybrid model combining 2D convolutional neural network (2DCNN) with VGG-16, featuring a deep architecture for extracting morphological characteristics, frequency features related to heart rate variability (HRV), and statistical attributes crucial for identifying atrial fibrillation (AF). Subsequently, to classify arrhythmia patterns, the VGG-16 Model is employed. Utilizing publicly available ECG image datasets, the proposed model achieved remarkable accuracy benchmarks: 99.61% accuracy, precision of 99.61%, and recall of 99.48%. Comparative analysis with existing approaches substantiates the efficiency and robustness of our model.

In order to improve the penetration ability of missiles in the end guidance stage, and considering the delay problem when designing the guidance control loop separately, this article mainly studies the design problem of an integrated guidance control law under input saturation and attitude angle constraints. For the lower triangular structure model of the integrated guidance and control system, this article mainly uses an improved backstepping method to design the controller, and introduces an error compensation term in the design of the virtual controller to improve system stability. Introducing auxiliary variables and barrier Lyapunov functions to handle saturation and state constraint problems; in the face of the main problems of atmospheric disturbances, unmodeled dynamic errors of the system, and unknown acceleration of the target, this article mainly adopts a finite time observer for processing. The final simulation results demonstrate the effectiveness and superiority of the controller designed in this article.

This paper presents an adaptive robust constraint controller for a continuous stirred tank reactor (CSTR) system based on a self-organizing fuzzy neural network (SOFNN). Due to the high complexity of the chemical reactions, the CSTR system contains many strong nonlinearities and uncertainties. This is the first time to introduce the SOFNN into the adaptive controller design of the CSTR system, which improves the adaptability to dynamic system changes through the adjustment of the fuzzy network structure. Meanwhile, the time-varying integral barrier Lyapunov functions (TVIBLFs) are employed to ensure the dimensionless reactant concentration and mixture temperature within a reasonable scope, which can improve the stability and safety of the CSTR system. Based on Lyapunov stability analysis, all the signals in a closed-loop system are ultimately bounded. Simulation results substantiate the efficacy of the proposed control scheme.

This paper introduces a novel hybrid approach, termed ZOA-SNN, for fault detection and identification in converter-interfaced microgrids. By integrating the Zebra Optimization Algorithm (ZOA) with Spiking Neural Network (SNN) technology, the proposed method provides a comprehensive solution suitable for both grid-connected and autonomous microgrid operation scenarios. The technique effectively isolates faults in the microgrid while maintaining operation continuity, particularly in islanded conditions. When operating in grid-connected mode, distributed generators (DGs) provide electricity as needed. When the grid is not available, power sharing amongst DGs is controlled by voltage angle droop control. By isolating malfunctioning portions, the proposed protection system reduces load shedding, while DG control guarantees smooth islanding and resynchronization. Evaluation on the MATLAB platform demonstrates the superior performance of the proposed technique compared to existing algorithms such as Augmented Lagrangian Particle Swarm Optimization (ALPSO), Graph Convolutional Network (GCN), and Buffalo Optimization (BO). With an accuracy, recall, precision, and F1-score reaching 98.5%, 99.2%, 99.1%, and 99.1%, respectively, the ZOA-SNN approach excels in fault detection and classification. Additionally, it significantly reduces computation times for parameter calculation, enhancing efficiency in microgrid control systems. These results highlight the innovation and advantages of the ZOA-SNN approach in enhancing the reliability and efficiency of fault detection systems in microgrid environments.

This article is aimed to study the parameter estimation problems of a non-commensurate fractional-order system with saturation and dead-zone nonlinearity. In order to reduce the structural complexity of the system, the model separation scheme is used to decompose the fractional-order nonlinear system into two subsystems, one includes the parameters of the linear part and the other includes the parameters of the nonlinear part. Then, we derive an auxiliary model separable gradient-based iterative algorithm with the help of the model separation scheme. In addition, to improve the utilization of the real time information, an auxiliary model separable multi-innovation gradient-based iterative algorithm is presented based on the sliding measurement window. Finally, the feasibility of the presented algorithms is validated by numerical simulations.

This article investigates the robust secure output tracking control for a class of uncertain two-dimensional (2-D) networked control systems. The 2-D systems are described by the well-known Fornasini–Marchesini (FM) local state-space model, the parameter uncertainties are assumed to reside in a polytopic region, and the deception attacks are supposed to occur randomly in the transmission process. In the problem formulation, Bernoulli random variables are used to characterize the phenomena of deception attacks. As main results, a novel $��{�H�}_{�\infty �}�$ performance analysis condition for the augmented closed-loop system is proposed by using a novel parameter-dependent Lyapunov function and some zero equalities. Furthermore, both parameter-dependent and parameter-independent controllers have been designed, respectively, such that the output of the 2-D systems tracks the output of a given reference model well in the $��{�H�}_{�\infty �}�$ sense. The design conditions are presented in terms of linear matrix inequalities (LMIs). In the end, a numerical simulation on a practical thermal process is applied to illustrate the effectiveness of the proposed method.

In this article, we propose an adaptive Kalman filtering with adaptive window length based on Q-learning for dynamic systems with unknown model information. The iteration step length of the Q-function is quantitatively adjusted through the influence function. The adaptive Kalman filtering algorithm is used to set an appropriate weight matrix for the Q-function to estimate unknown model parameters. One numerical example and a practice-oriented case are given to illustrate the effectiveness of the proposed method. It is shown that this filtering can provide state estimates of best accuracy among all the compared methods when the model mismatch and noise statistical characteristics change.

This article presents a novel multi-thread attracting manifold (MTAM) adaptive control scheme. The novel contribution of this new controller lies in how the attracting manifold design method is spread across multiple estimates of the truth while simultaneously weighting between said individual estimates to arrive at an adaptive composite estimate which outperforms single thread adaptive controllers. A notable benefit of this new formulation is that it allows for both the weighting of multiple threads and monotonic decrease of parameter error to aid the performance of transient error dynamics, without arbitrarily increasing adaptation gains. The benefits of the proposed controller over existing adaptive control schemes are analyzed through two aerospace flight control applications from the literature.