ISA transactions最新文献

This paper proposes an adaptive PID-based sliding mode controller (APID-SMC) for autonomous underwater vehicles (AUVs), optimized using ant colony optimization (ACO), to enhance trajectory-tracking accuracy and robustness under external disturbances. The proposed controller demonstrates significant improvements over conventional SMC, STA, and PID controllers across multiple performance indices. Specifically, the APID-SMC reduces the integral absolute error (IAE) in the surge, sway, and yaw channels by 14.50%, 27.97%, and 26.39%, respectively, and improves ITAE by 66.80%, 80.17%, and 82.84%, highlighting its superior transient performance. The controller also generates smoother control signals with reduced chattering and maintains stability under extreme noise and uncertainties. The framework integrates the robustness of sliding mode control with the smooth corrective action of a PID controller, whose gains are dynamically tuned online via a gradient descent algorithm (GDA). Additionally, ACO optimally selects learning rates and sliding surface coefficients by minimizing a trajectory-tracking cost function, ensuring rapid convergence and consistent performance. These results confirm that APID-SMC is a highly effective and practical control solution for complex and uncertain marine environments.

This study proposes an integrated framework to improve indoor air quality (IAQ) resilience and ventilation reliability in enclosed subway environments. Fault tree analysis and Monte Carlo simulations were used to identify critical vulnerabilities, supporting system-level fault prognostics. Three targeted strategies were developed: mechanical redundancy (MRS), health-aware feedback control (CRS), and data-driven fault detection and reconstruction (DRS). These strategies were experimentally and computationally validated, both individually and in combination. While standalone strategies offered partial improvements, the integrated application achieved the greatest enhancement, increasing the health-resilient ventilation index (HRVI) by 21.86% in normal and 23.31% in sensor fault conditions. The results highlight the necessity of multidimensional approaches that combine prognostics, health-aware control, and data assurance to ensure resilient IAQ management in complex subway systems.

An adaptive fractional-order terminal sliding mode control technique is presented in this paper to improve quadrotor UAV control performance in the presence of stochastic disturbances and uncertainties. In order to fulfill mission-critical timing limitations, a fractional-order sliding mode controller is incorporated into the finite-time theory to guarantee system stability within a finite period. Second, a super-twisting sliding mode observer is integrated to estimate external disturbances, which enhances the system's resilience to environmental uncertainties, wind gusts and aerodynamic impacts. The Lyapunov theory is conducted to validate the closed-loop stability of the suggested controller. Conclusively, numerical simulations and semi-physical experiments demonstrated the effectiveness of the synthesized method with the existing results.

In practical cooperative aerial transportation, the system is inevitably exposed to unmodeled and unpredictable disturbances, such as wind gusts, internal interaction forces arising from agent-load coupling, and inaccuracies in load mass information, making robustness and adaptive disturbance rejection essential for reliable real-world deployment. This paper presents an adaptive, robust trajectory-tracking control strategy for a team of unmanned aerial vehicles collaboratively transporting a cable-suspended load. To assess the method's effectiveness, extensive real-world experiments were conducted across multiple scenarios, including conditions with unforeseen disturbances. Experiments involving up to five aerial agents, combined with external disturbances and load-parameter uncertainties, further demonstrate the robustness and scalability of the proposed approach, while a formation-reconfiguration experiment highlights its adaptability during task execution. The results demonstrate that the proposed controller ensures accurate trajectory tracking of the suspended load and consistently outperforms two benchmark controllers with distinct characteristics, achieving up to 83% performance improvement in certain scenarios. These findings highlight the robustness and applicability of the method for practical deployment.

Because electric vehicles (EVs) are a clean energy source, understanding battery aging and health is a hot research topic. Such evaluations are essential for timely maintenance, ensuring operational efficiency, and enabling second-life applications of lithium-ion battery (LIB) packs. This study presents a new framework for predicting capacity loss in EV-compatible LIB packs using a quantum long short-term memory (QLSTM) neural network with a shared linear embedding layer, transfer learning, and a sliding window technique. The method is tested with five LIB pack datasets collected over 29 months from charging devices. Results show the approach's robustness, providing improved accuracy and efficiency, especially when trained on multiple battery datasets. Models that combine QLSTM with linear embedding layers can predict capacity with a root-mean-square error (RMSE) of less than 0.003 for the target LIB packs. These findings highlight the framework's potential to improve predictions of capacity degradation and to support sustainable energy management in EV batteries.

This study proposes a soft-sensor-based method to significantly shorten compressor performance evaluation tests and presents the results of its industrial application over five years by a compressor manufacturer. Traditional approaches demand long testing times to reach steady-state conditions, with overall test durations frequently surpassing two hours. In this work, a soft sensor based on a time-delay neural network (TDNN) was developed to monitor steady-state conditions of key performance parameters - cooling capacity, power consumption, and coefficient of performance - and to predict their final values. A dataset of 392 compressor evaluations was used for model development, and the proposed method achieved an overall reduction in test duration of close to 50%. This is accomplished because the proposed method incorporates a soft-sensing approach, trained on historical data, facilitating early detection of steady-state conditions and accelerating testing procedures. During five years of real industrial application, with the proposed approach tested in 9184 performance evaluations, this method demonstrated a 55% improvement in total test time, with more than 95% of the tests showing prediction errors below 2%. Therefore, the proposed tools resulted in consistent time savings and increased operational efficiency during their industrial evaluation.

Under the multi-variety and small-batch production mode, the machining parameters switch segments throughout the tool life cycle. This segmented parameter switching leads to frequent jumps in degradation indicators at the transition points between segments. Furthermore, existing methods rely on a single degradation indicator for degradation modeling, which is insufficient to comprehensively reflect the tool degradation process. These issues significantly impede accurate tool remaining useful life (RUL) prediction under segmented variable machining conditions. Therefore, a method based on condition-adaptive degradation indicator and binary Wiener process is proposed for milling tool RUL prediction. The method initially extracts two highly characteristic degradation indicators from multi-source sensing signals. Subsequently, the effect of segmented variable machining parameters on these indicators is eliminated through a baseline transformation algorithm. Then, a nonlinear binary Wiener process with random effects is developed to depict the correlation between the degradation indicators. Besides, the Copula function is employed to model the dependence between the marginal RUL distributions of the two degradation indicators, thereby predicting the RUL under the coupling of degradation indicators. Finally, a segmented variable-condition milling experiment is carried out to validate the proposed method.

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.