2026-11-10 2026, Volume 10 Issue 4

  • Select all
  • research-article
    Meiqi Yuan, Jun Cai, Xinyu Luo, Ying Lu, Tinghao Ren

    The disorderly integration of small hydropower plants into distribution networks can easily lead to problems such as curtailed power generation, voltage overshooting, and difficulty in regulation. Therefore, it is crucial to study small hydropower clustering methods for regional coordinated control and optimal scheduling. Traditional methods often rely on geographic location or administrative divisions for clustering, but they struggle to simultaneously account for multidimensional characteristics such as electrical characteristics, operational characteristics, and spatial distribution of power plants, thus failing to meet the optimized regulation requirements of distribution networks. To this end, this paper introduces complex network analysis methods and proposes a small hydropower cluster partitioning method based on the multidimensional feature network model and the community detection algorithm. Hydropower stations are modeled as network nodes, different coupling dimensions are mapped into weighted edges, and the similarity network fusion (SNF) method is employed to achieve multidimensional feature fusion. Furthermore, a hydropower coordination index is defined to quantify cluster quality in multiple spatiotemporal scenarios. Clustering is then performed using the Louvain community detection algorithm. Small hydropower stations in the Meijiang River Basin, Zunyi, China, were selected as the case study. The results show that the proposed method can improve the physical interpretability of the partitioning results and provide an effective solution for the coordinated control and optimized regional scheduling of small hydropower stations.

  • research-article
    Kristian Gochi, Ahtziri González, Juan Anzurez

    We propose a persistent homology procedure to estimate the severity of mixed eccentricity in three-phase induction motors from line-current measurements. Currents are generated in ANSYS RMxprt/Maxwell under no-load operation for both star and delta connections across six eccentricity levels. Each steady-state frame is embedded as a point cloud in R3 and approximated with alpha complexes to compute persistence diagrams and Betti curves. We define descriptors based on the topological properties and behaviors of the Betti curves. Under the delta connection, we add a new descriptor associated with noise in 2-homology. All descriptors exhibit monotonic trends with increasing eccentricity and allow quadratic fits that enable interpolation of intermediate eccentricity levels. We also show that alpha complexes implemented in GUDHI provide substantial computational advantages over Vietoris-Rips complexes implemented in Ripser.

  • research-article
    Hoai-Thuong Luong, Thanh-Tung Pham, Minh-Thanh Le, Chi-Ngon Nguyen

    This paper presents a rigorous experimental evaluation of a particle swarm optimization (PSO)-optimized linear quadratic regulator (LQR) controller deployed on a physical Quanser 3-degree-of-freedom (3-DOF) gyroscope platform. To address the limitations of manual trial-and-error parameter tuning under real-world constraints, a multi-criteria optimization framework was established using three distinct performance indices: integral of absolute error (IAE), integral of time-weighted absolute error (ITAE), and integral of squared error (ISE). Real-time hardware experiments under multi-frequent square-wave and continuous sinusoidal trajectories, alongside impulse disturbance scenarios, demonstrated that the proposed PSO-LQR configured via the IAE criterion achieves the most superior balance between dynamic tracking agility and actuator energy efficiency. Specifically, the proposed PSO-IAE framework yields a settling time of 1.68 s, zero overshoot, and a minimal steady-state error of 0.10°. Crucially, quantitative energy analysis revealed that the proposed framework effectively circumvents motor voltage saturation, operating at an optimal control torque root mean square (RMS) value of 0.0962 Nm, which significantly outperforms the ISE configuration (0.3184 Nm) and the ITAE variant (0.1446 Nm). These empirical findings provide valuable practical benchmarks for constraint-enforced automation in high-level mechatronic mechanisms.

  • research-article
    Abdelwahhab Khaldi, Choukri Bensalah, Amal Choukchou Braham

    Existing solutions to attitude estimation for unmanned aerial vehicles (UAVs) in GPS- denied environments face a persistent trade-off: classical filtering approaches such as the extended Kalman filter (EKF) suffer from linearization errors, while geometric nonlinear observers on the special orthogonal (SO(3)) Lie group avoid this issue but rely on a manually tuned, fixed correction gain. A fixed gain cannot simultaneously deliver fast convergence during nominal operation and robust noise rejection when inertial sensors degrade, and existing adaptive or gain-scheduled alternatives either require manual heuristics or lack a real-time responsiveness mechanism, leaving gain tuning an open and recognized limitation in this line of work. This paper proposes a robust vision- inertial fusion method for UAV attitude estimation using a nonlinear observer on the SO(3) Lie group with adaptive gain control that directly addresses this gap. We introduce a PID-based gain adaptation mechanism that dynamically modulates the observer’s correction term based on real-time gyroscope variance metrics. When inertial data deteriorates due to noise or bias, the observer automatically relies more on visual corrections; during stable periods, it prioritizes inertial data for faster convergence. MATLAB/Simulink simulation results demonstrate that the adaptive observer reduces attitude estimation error by approximately 65 percent during disturbances and achieves reconvergence over 10 times faster after disturbances compared to fixed-gain baselines, while maintaining bounded errors during severe gyroscope perturbations where fixed gain observers diverge. This work provides a computationally efficient solution to the gain-tuning challenge in geometric attitude estimation for autonomous UAVs operating in GPS-denied environments. Validation is performed through comprehensive simulation studies.

  • research-article
    Ying Foo Leong, Nur Huda Ramlan, Suliana Ab-Ghani, Hamdan Daniyal, Mohd Herwan Sulaiman

    Off-board fast-charging systems for electric vehicles require high-power DC-DC converters capable of maintaining rapid transient regulation without overshooting current boundaries. Dual Active Bridge converters satisfy these topology requirements, but conventional Proportional-Integral controllers tuned via static methods struggle under severe load steps and non-linear battery dynamics. In this study, we present a Dynamic Barnacles Mating Optimization framework engineered to automate controller gain synthesis. By introducing a time-varying linear decay mechanism to regulate the mating search radius, the proposed algorithm resolves the exploration-exploitation imbalance inherent in standard metaheuristic search strategies. Algorithm performance was benchmarked across 23 classical test functions, followed by offline converter simulation models and real-time validation on a hardware emulator. Statistical analysis under a Holm-Bonferroni corrected test confirmed that the dynamic search mechanism significantly improves solution accuracy and stability over standard barnacles optimization without incurring statistical losses. When applied to controller tuning, the proposed optimizer effectively suppressed transient current overshoot during severe load steps in offline simulation models. In real-time hardware emulation, the controller suppressed supply voltage disturbances within 7.2 milliseconds and restricted startup current overshoot below 3.5%. These results verify that dynamic search radius decay synthesizes well-damped control parameters, preventing overcurrent protection tripping and ensuring safe, stable electric vehicle fast-charging.

  • research-article
    Hai Pham Hong, Tung Le Duc, Minh Tran Anh

    This study presents an enhanced three-dimensional conjugate heat transfer (CHT) model designed to improve the thermal simulation accuracy of a 320 kVA, 22/0.4 kV cast-resin dry-type transformer (CRDT). While conventional thermal modeling commonly assumes constant thermophysical properties for epoxy encapsulation, this research dynamically incorporates temperature-dependent functions for thermal conductivity (λ) and specific heat capacity (𝐶𝑝). These parameters, previously identified through experimental measurements (ASTM D5334) on the specific 8055A/8055B epoxy system, are continuously updated at every solver iteration according to the local temperature field. By utilizing a one-way multiphysics coupling strategy, the non-uniform electromagnetic losses computed via ANSYS Maxwell are explicitly mapped as volumetric heat sources into the ANSYS Fluent CHT model. The proposed framework was rigorously validated against empirical data collected from 27 thermocouples installed on an actual industrial prototype operating under rated load conditions. The integration of temperature-dependent material properties significantly reduced the maximum relative prediction error from 10.9% (using conventional constant properties) to 4.59% along the axial profile of the high-voltage winding. Furthermore, the error margin remained below 5.1% across all 27 validated locations, successfully pinpointing the hotspot temperature of 75.0°C at the upper section of the phase-B low-voltage winding. Given that the thermal aging of epoxy insulation accelerates exponentially with rising temperatures, this substantial improvement in prediction accuracy provides a highly credible foundation for estimating insulation service life and establishing safer, better-founded design margins for CRDTs.

Publishing model
0

Downloads

Submission to first decision (median)
41 days

ISSN 2578-1588 (Online)