Advanced Control for Applications最新文献

Considering the limited prediction accuracy of the existing time prediction model, we propose a bus arrival time prediction model based on improved SVM and H∞ filter technique. The NN-SVM which had two input features was used to predict the baseline of bus running time from historical trip data. Applying the real-time running information and combining it with baseline time, we use H∞ filter to predict bus arrival time dynamically. Bus arrival time forecasted by the proposed model was assessed with the data of transit route number 142 in Changsha in China. Results demonstrate that the proposed method achieved a Mean Absolute Error (MAE) of 19.47 s, Root Mean Square Error (RMSE) of 23.42 s, representing a 65.0% reduction in MAE and a 74.5% reduction in RMSE compared to standard SVM. and PBIAS of −2.4%, indicating high accuracy with minimal underestimation bias. Compared with conventional SVM and Kalman-based models, the proposed model reduced RMSE by 19.4% and improved robustness in dynamic traffic scenarios.

In this article, a novel mathematical model of a quadrotor UAV (Unmanned Aerial Vehicle) suffering from one type of structural fault, angle deflection in a motor and the corresponding mechanical link, has been derived. Such faults add additional nonlinear terms to the differential equations of motion and change the 3D configuration of the UAV. These nonlinear terms rely significantly on the unknown fault angles, which are estimated via a Radial Basis Function Neural Network (RBFNN). Moreover, a Non-singular Fast Terminal Sliding Mode Control (NFTSMC) scheme optimized by a Genetic Algorithm (GANFTSMC) has been designed for trajectory tracking and reduction of control effort. Simulation results for the entire closed-loop system, using three different types of reference signals and fault angles, demonstrate the significant performance of the proposed controller in the presence of structural faults and external disturbances. Furthermore, the settling time of error dynamics of the system states to the origin has been finite, and the superior performance of our proposed control strategy has been validated via comparison with other robust nonlinear control techniques implemented in the literature.

The increasing penetration of nonlinear loads and power electronic devices in distribution networks has significantly intensified power quality (PQ) challenges, including harmonic distortion, voltage sags/swells, and reactive power imbalance. This paper proposes advanced intelligent control strategies for a Distributed Static Compensator (DSTATCOM) integrated with a Permanent Magnet Synchronous Generator (PMSG)-based Wind Energy Conversion System (WECS) and Battery Storage System (BSS). The proposed schemes include (1) a hybrid fuzzy-sliding mode control (HFSMC), (2) an artificial neural network-based PI controller (ANN-PI), and (3) a Butterfly Generator-based particle swarm optimization PI controller (BG-PSO-PI) for Maximum Power Point Tracking (MPPT) and real-time voltage/current compensation. Simulation results under nonlinear load and fault conditions demonstrate that the proposed BG-PSO-PI controller achieves a minimum source current Total Harmonic Distortion (THD) of 3.51%, outperforming ANN-PI (3.57%), Fuzzy-SMC (3.72%), and conventional PI (7.03%) controllers. Furthermore, the BG-PSO-PI controller maintains DC-link voltage deviations within ±2%, compared to ±5% in traditional schemes, and achieves fault recovery in less than 0.05 s during grid disturbances. These results confirm that the proposed control architecture significantly improves PQ indices, enhances voltage stability, and ensures rapid dynamic response, making it suitable for modern grid-integrated renewable energy systems.

The existing spatial motion control methods for industrial production robotic arms suffer from sluggish convergence efficiency and inadequate control accuracy. With the aim of resolving this matter, the research proposes an optimization method for spatial motion control of robotic arms based on modeling analysis and the fast extended random tree algorithm. The study first models and analyzes the spatial motion of the robotic arm, and combines the Monte Carlo method for random sampling to create a significant amount of feasible paths. The difference between the feasible path and the planned path is that the feasible path is optimized through multiple iterations, gradually approaching the optimal solution, while the planned path is a predetermined fixed path. Afterwards, the improved fast expansion random tree algorithm is used to optimize the path and achieve efficient planning of the robotic arm path. Finally, the proportional integral derivative control method and the recursive least squares method are employed to flexibly control the spatial motion end of the robotic arm, improving the motion accuracy and stability of the robotic arm. The experimental outcomes reveal that the node utilization rate and planned path length of the proposed control method in simulation analysis are 20.71% and 111.61 cm, respectively, and the fitting degree between its control trajectory and the ideal trajectory exceeds 0.98. In practical applications, the average time required for the robotic arm using the proposed method is only 12.5 s, and the quality pass rate remains above 98%. In addition, the completion rate of its winding task is 99.5%. The proposed control optimization approach can effectively improve the control accuracy and production quality of industrial production robotic arms.

This study presents a robust control strategy for unstable cascade processes with time delays, integrating Particle Swarm Optimization (PSO) and the Internal Model Control (IMC) framework. The proposed methodology employs a dual-controller architecture, where the secondary controller is systematically tuned using the IMC approach, while the primary controller is optimized via PSO to enhance closed-loop stability and performance. Extensive simulation studies were conducted across various unstable cascade processes to evaluate the effectiveness of the proposed approach in both servo and regulatory tasks. Comparative analysis with state-of-the-art control methodologies demonstrates that the proposed strategy achieves superior closed-loop performance, particularly in handling system uncertainties. A comprehensive numerical assessment using multiple performance indices indicates a substantial reduction in total error by 65% and control effort by 43%. The evaluation metrics include error minimization, total variation, rise time, and overshoot percentage, affirming the efficacy and robustness of the proposed control scheme.

This article presents the adaptive sliding mode method for the gun elevation system of the tank operating under model uncertainties and external disturbances. The proposed controller combines optimal control design, the adaptive compensation mechanism using the radial basis function (RBF) neural network, while integrating the sliding mode control (SMC) law to enhance robustness and trajectory tracking accuracy. The RBF network is used to estimate and compensate for unknown nonlinear components and dynamic uncertainties in real time, and the SMC law is incorporated to ensure robustness and force the system output to accurately follow the desired trajectory. The control strategy is synthesized to meet strict performance requirements under complex real-world operating conditions. Simulation studies conducted in Matlab evaluate the controller's effectiveness. The results demonstrate that the proposed method achieves accurate trajectory tracking, strong disturbance rejection, and improved robustness, confirming its potential for practical military applications.

Inverted pendulum is an example of a classical problem in control theory that has been widely used for investigating control algorithms like state feedback, artificial neural networks, fuzzy control, and robust control. The piecewise Affine-fuzzy (PWA-Fuzzy) approximation of an inverted pendulum on a cart has been investigated in previous literature. The main drawbacks of PWA approximation, i.e., discontinuity in control signal and chattering between different regions, are addressed by the PWA-Fuzzy model. In this paper, with the aim of stabilizing the system in the open-loop unstable equilibrium point, a patched-fuzzy state feedback (PFSF) controller is designed as an improved form of the conventional state feedback controller for the PWA-Fuzzy model of an inverted pendulum on a cart. Because of the intention to implement the mentioned controller for a real plant, the identification of PWA-Fuzzy model parameters by the linear least squares method based on the numerical method is presented. Furthermore, the implementation of the mentioned controller using two different techniques, including analog and digital circuits, is presented. Finally, in order to evaluate the proposed method, the simulation and experimental results are compared.

Oscillatory behavior in process control loops is a persistent challenge in industrial plants, often resulting in diminished control performance, increased energy consumption, and economic losses. If left unaddressed, such oscillations can propagate throughout the plant, causing variability in downstream processes and negatively impacting throughput and product quality. Accurate detection and classification of oscillations, along with identification of their root causes, are therefore critical for enabling timely corrective actions that enhance control performance and overall process efficiency. Common causes of oscillations in process control include control valve stiction, suboptimal PID tuning, measurement noise, and external disturbances, each imparting distinct dynamic patterns on the process variable (PV) and controller output (OP). Manual detection and classification of these oscillations through visual analysis is time-consuming and impractical due to the large number of control loops in modern plants. In this paper, we present an automated deep learning framework for the detection and classification of oscillations in process control loops. The proposed method employs a one-dimensional convolutional neural network (1D-CNN) to analyze time-series data from PV and OP signals, enabling the model to learn and distinguish between different oscillation patterns associated with various root causes. The framework is trained and validated using both simulated datasets and real industrial plant data, ensuring robustness across a wide range of oscillation scenarios. Case studies are provided to illustrate the practical application of the method, and results demonstrate that the proposed approach achieves high accuracy in both detecting the presence of oscillations and correctly identifying their underlying causes. This automated solution offers a scalable and efficient tool for plant operators and engineers, supporting proactive maintenance, improved control loop reliability, and enhanced overall plant performance.