ABSTRACT The issue of eliminating the chattering phenomenon and lessening the sliding control controllers' convergence time for speed control of induction machines (IMs) is a fundamental challenge. This paper addresses this issue and presents a sliding mode controller that has a short convergence time and negligible chattering. Barrier function‐based adaptive fast terminal sliding mode (TSM) control (BFBAFTSMC). The reaching law of this controller guarantees that the tracking error will eventually converge to a finite value under any initial conditions. This barrier function feature of the controller eliminates the chattering phenomenon and increases resistance to disturbances. On the other hand, this controller uses an adaptive fast TS surface that decides that the convergence time is very limited. Therefore, it can be concluded that this controller has high robustness, high convergence speed, and low chattering. An induction motor was the subject of a number of useful laboratory tests to examine the effectiveness of this approach. The findings demonstrate that it is successful in decreasing the chattering phenomena and convergence time while additionally strengthening the system's resilience.
ABSTRACT The problem of model uncertainty and the variation of modeling parameters of the motor is one of the key issues facing induction machine drives. Hence, the control strategy of this class of motors should cope with uncertainties and finally deliver accurate control results. However, on the other hand, convergence time is another issue in control strategies of induction machines, so that with the variations of the reference speed, the controller should track the reference signal and make the motor speed reach it quickly. This paper proposes the prescribed sliding mode control (PSMC) technique. Precisely, based on the Lyapunov theory, stability of the system will be proved in a specified period of time, so that motor speed reaches the reference speed within a specific time. This way, we can ensure that the controller convergence time can be predetermined and achieved before designing this controller. In addition, the control loop's resistance to changes in parameters or model uncertainties will also be resolved. In order to examine the performance of the proposed method, a series of practical experiments were conducted on an induction motor in the laboratory, and the results obtained confirm the effective performance of this method in reducing convergence time and resistance to uncertainties. The results of practical experiments show that the proposed method has reduced the convergence time to 0.1 s and the steady‐state error to 0.9 rpm, which is a significant improvement over conventional methods such as ASMC and PI. Also, the robustness of the proposed method against parameter variations of up to ±10% has been confirmed with an error of less than 2.5 rpm, indicating its effective performance under conditions of model uncertainty.
ABSTRACT Real‐time power grid dispatching (PGD) is an essential task in power‐system operation, but it remains challenging because of high dimensionality, nonlinearity, renewable‐energy uncertainty, and strict response‐time requirements. Classical optimization‐based methods may become computationally burdensome or insufficiently adaptive under rapidly changing renewable generation. Existing multi‐agent reinforcement learning (MARL) strategies provide a promising data‐driven alternative, yet their practical use is limited by coordination blindness under partial observations, weak semantic understanding of grid topology and fault evolution, and limited generalization under rare events. To address these issues, this paper proposes a large‐language‐model‐powered agent‐orchestration framework, termed LLM‐AOM, for real‐time PGD. The framework contains a high‐level orchestration layer and a low‐level execution layer. The LLM is used only as a safety‐filtered supervisory orchestrator: it converts grid states into structured situational‐awareness descriptions and issues macro‐level coordination guidance. Physical dispatch actions are generated by MAPPO‐based low‐level agents within predefined operational limits. For reproducibility and feasibility, the implementation uses Grid2Op‐based simulation, an IEEE 39‐bus test system, structured LLM‐output parsing, bounded reward shaping, safety filtering, fallback execution, latency measurement, and matched‐seed comparison. Simulation results on the IEEE 39‐bus benchmark indicate that LLM‐AOM improves security‐related indicators and post‐contingency recovery relative to DCOPF, PPO, and MAPPO in the tested scenarios. The study therefore provides a proof‐of‐concept for using LLMs as interpretable, safety‐filtered high‐level orchestrators rather than direct physical controllers in critical infrastructure control.
ABSTRACT Specialized electrical infrastructure of Traction Power Supply Systems (TPSS) adopts the high voltage levels while varying load current dynamics, which is distinct from a general power system by handling the high power and train loads. However, transformer saturation, unbalanced loads, non‐linear power electronic devices, and system resonance factors introduced the harmonics in the TPSS, which resulted in equipment damage issues and contributed to performance degradation. Consequently, the research work implements the Intelligent Swooping Optimization‐based Q‐Learning distributed one‐dimensional Convolutional Network (ISO‐QLCNet) model to perform the harmonic mitigation task in the TPSS. The QLCNet model predicts potential failures and learns the optimal control policies to mitigate the harmonics in the railway operations. Moreover, the developed model utilized the dynamic and non‐linear loads to identify the complex harmonic patterns and indicate the power quality (PQ) issues to achieve the mitigation task effectively. Furthermore, the model parameters are optimized by the ISO algorithm, which helps to meet the maximum energy efficiency and lower total harmonic distortion (THD). Eventually, the ISO‐QLCNet model exhibits the THD, train‐1 voltage, train‐1 current, switching output, train‐2 current, and train‐2 voltage are 0.62, 562.26 V, 8217.19 A, 5260.74, 961.54 A, and 632.49 V, respectively.
ABSTRACT This academic article introduces a wind system (WS) that uses a parallel arrangement of five‐phase permanent magnet synchronous generators (PMSGs). The control mechanism for this system uses a fifteen‐switch rectifier (FSR) architecture, which has been particularly designed for grid‐connected applications. To enhance the control performance of the WS being suggested, the backstepping control (BSC) approach is used for both the converters on the generators side and the converters on the grid side. The efficacy of the suggested control technique is assessed by conducting simulations under different wind speeds and comparing it to a control system that utilizes a proportional–integral (PI) controller. The simulation findings confirm that the suggested control mechanism guarantees precise monitoring of the regulated variables. Furthermore, significant enhancements are seen in several facets of the system's performance. More precisely, the suggested control technique decreases active power fluctuations by 10% and reactive power by 20%, limits the increase in DC bus voltage to 4.4%, and speeds up overshoot by 18%. The network current's total harmonic distortion is 41.8% lower compared to the traditional approach. In addition, the system's efficiency has been improved to 96%, exceeding the 94.3% efficiency attained using the backstepping approach. The suggested technique demonstrates higher performance when compared to a conventional method.
