Transportation electrification is a cornerstone in addressing climate change, primarily through the adoption of electric vehicles (EVs), which significantly reduce greenhouse gas emissions. Among various electric motor technologies, switched reluctance motors (SRMs) have emerged as promising alternatives due to their simple design, fault tolerance, and robustness. However, challenges such as torque ripple, high acoustic noise, and efficiency limitations hinder their widespread adoption. This paper presents a comprehensive review of contemporary SRM control strategies and associated power converters, aimed at improving the performance of EV applications. The study explores fundamental electromagnetic principles, highlights torque control strategies (e.g., indirect torque control, direct torque control, and artificial intelligence-based torque control), and evaluates their efficacy in minimizing torque ripple and optimizing motor performance. Additionally, the paper assesses various power converter topologies, emphasizing asymmetric half-bridge, novel integrated power converter, and T-type converters for their suitability in EV systems. Based on an extensive review, a four-phase SRM driven by a T-type converter, coupled with direct instantaneous torque control, is identified as the optimal configuration for EVs, providing a cost-effective, reliable, and high-performance solution for sustainable transportation.

In this paper, we apply an α-tension field from differential geometry to classical image processing tasks of denoising with edge preservation and multiphase feature enhancement. The main contribution of this work is that it is the first systematic investigation of the α-tension field for image processing. Contrary to traditional operators, such as the Laplacian, which are susceptible to noise amplification or are ineffective for complex structures, the α-tension field relies on a nonlinear adaptive mechanism depending on the magnitudes of local gradients. It allows effective denoising and retains edges and fine details by utilizing higher-order gradient information. The field of α-tension provides more sensitive and adaptive models than linear models, such as total variation regularization, anisotropic diffusion, etc. The study exemplifies its advantages over previous methods in preserving structural integrity and minimizing artifacts. It also considers numerical implementation issues and provides guidelines for real-time and large-scale processing. This framework adds up to the known need for faster image-processing tools while links connections to differential geometry.

The article addresses the solution of parabolic differential equations with integral boundary conditions using the Haar wavelet collocation method. This approach employs a linear combination of Haar wavelet functions to estimate the largest derivatives in the governing equation. The integral boundary conditions are incorporated by repeatedly integrating the highest derivative to formulate equations for the unknowns. Haar wavelets are particularly suitable for approximating solutions to differential equations due to their compact support and multiresolution properties. Numerical experiments on various test cases show that the proposed method yields accurate results, especially when the parameters of the integral boundary conditions are negative.

Uncertain incommensurate fractional nabla difference systems (IFDSs) in recurrent neural networks (RNNs) are analyzed using fuzzy number theory to address input uncertainties. Fuzzy number theory and its operations are re-investigated, and the H-differenceable concept is introduced. The existence of a unique H-differenceable solution for incommensurate RNNs is proved. A recursive algorithm is proposed to obtain fuzzy solutions. Illustrative examples with 2-dimensional IFDSs are provided to validate the framework for integrating fractional calculus, fuzzy dynamics and incommensurate RNNs.

This study proposes an optimal control strategy for battery energy storage systems to support frequency regulation in power systems with high renewable energy penetration. The algorithm generates a sinusoidal reference signal for the pulse-width modulation scheme of the direct current/alternating current converter, adjusting it based on voltage magnitude and phase set points. The control system integrates multiple loops to manage frequency, voltage, active and reactive powers, charge, and current controllers on the d- and q-axes. It limits frequency deviations and improves the rate of change of frequency. An adaptive nonlinear droop control method, combined with state-of-charge (SOC) feedback, regulates active power-frequency control to enhance grid stability. The SOC feedback mechanism enables dynamic charge and discharge operations, ensuring that frequency remains within operational limits. Simulation results, validated using modified Vietnamese Tay Nguyen 500/220 kV and IEEE 39-bus systems with DIgSILENT/PowerFactory, show that the proposed method outperforms the conventional CBEST model in scenarios involving sudden generation outages or fluctuating renewable energy output. This method meets the frequency stability requirements set by the Circular No. 25/2016/TT-BCT of the Vietnamese Ministry of Industry and Trade, ensuring that the system operates within a stable frequency range of 49.5-50.5 Hz under 120 s and recovers to 49.8-50.2 Hz within 300 s.

Accurate fault diagnosis of rolling bearings is hindered by the weak nature of early fault signals and the limited availability of labeled data, especially under small-sample conditions. To overcome these challenges, this paper proposes a novel method combining golden jackal optimization (GJO) with improved variational mode decomposition (VMD), enhanced feature extraction, and optimized classification. First, GJO is used to optimally determine the key decomposition parameters of VMD, thereby improving the accuracy of vibration signal decomposition. A comprehensive discrimination factor algorithm then selects fault-sensitive intrinsic mode functions, and the signal is reconstructed to enhance fault characteristics. Multiscale fuzzy entropy is calculated from the reconstructed signals at multiple scales to build distinct state feature vectors. These vectors are fed into a back-propagation neural network optimized via the golden sine subtraction-average-based optimizer for precise fault classification. The method’s effectiveness is verified through simulation and experimental data. Compared with conventional approaches, it shows superior performance in extracting weak fault features and maintaining high diagnostic accuracy under small-sample scenarios. This integrated framework presents a robust solution for rolling bearing fault diagnosis.

This article presents a novel control approach for robust fixed-time trajectory tracking in nonlinear dynamic systems affected by external disturbances and model uncertainties, utilizing a fixed-time disturbance observer. Initially, a new fast disturbance observer was designed to reliably estimate external disturbances and model uncertainties within a fixed timeframe, independent of initial conditions and without requiring strict assumptions about the nature of these disturbances and uncertainties. Based on the disturbance estimates, a new robust fixed-time trajectory tracking sliding mode control strategy was developed, incorporating a fixed-time sliding variable and a reaching law with a state-dependent exponent coefficient. Using Lyapunov-based analysis, it is proven that the tracking errors of the closed-loop system converge to a neighborhood of the origin within a fixed time, independent of the initial conditions. Finally, comprehensive simulations were conducted to validate the effectiveness of the proposed strategy, demonstrating its advantages in achieving fast convergence, avoiding singularities, reducing chattering, and compensating for model uncertainties and external disturbances.

Based on data from 40 Specialized and New Enterprises (SNEs) in Zhejiang from 2017 to 2021, green innovation efficiency (GIE) is assessed by Charnes-Cooper-Rhodes and super-Slack-Based Measure models, and influencing factors of GIE are analyzed using the Systematic Generalized Method of Moments Dynamic model to address the improvement of GIE in SNEs. The conclusions include: (i) the GIE is declining from 2017 to 2021 and could be improved in the future, especially for SNEs in Zhejiang. (ii) As for influencing factors, research and development investment, industry-university-research collaboration, and government support have positive effects, while enterprise scale has a negative effect of restraining the development of green innovation of SNEs in Zhejiang. Therefore, several countermeasures, including establishing sound scientific research mechanisms, forming cooperation mechanisms among enterprises, research institutes, and colleges and universities, strengthening government support, and improving the government’s subsidy policy, are being put forward. Scientific basis and practical guidance are provided to enhance the green innovation capability of enterprises and the formulation of relevant policies by the government.

Rigid robotic manipulators encounter several challenges in trajectory tracking control, including low accuracy and poor stability, resulting from uncertainties, external disturbances, and parameter variations. To address these issues, this study proposes two hybrid controllers that integrate the strengths of proportional-integral-derivative (PID) control with neural network (NN) methods for a three-link rigid robotic manipulator. These hybrid structures are the NN-PID controller and the self-tuning NN with PID (STNN-PID) controller. Their performance is compared against that of a conventional PID controller. To optimize control performance metrics, such as the integral time square error (ITSE), the parameters of the proposed controllers were tuned using the African vultures optimization algorithm. MATLAB was used to evaluate the effectiveness. Robustness tests were performed by varying the initial conditions, introducing external disturbances, and modifying system parameters. The NN-PID controller achieved ITSE values of 0.. 28919 × 10⁻⁴, 0.064321, and 0.001164, respectively, while the STNN-PID controller yielded values of 3. 54549 × 10⁻⁴, 3.526199, and 0.883710, respectively. Moreover, when all these conditions were applied simultaneously, the NN-PID controller achieved an ITSE of 0.073968, compared to 2.672754 for the STNN-PID controller. These results demonstrate that the NN-PID controller outperforms the other controllers across all testing conditions. These findings confirm that the NN-PID controller is the most effective controller in terms of tracking accuracy, stability, and robustness across all test scenarios.

In this paper, model-reference adaptive control (MRAC) with neural network (NN) and time delay estimation (TDE) is proposed for controlling a robotic manipulator. With more than two degrees of freedom (DoF) of the robot, the formulation of a known regression matrix is tedious and also difficult to compute for the different robotic systems. Therefore, this work introduces MRAC based on TDE with NN (MRAC-NNTDE) to achieve high-control performance without prior knowledge of the regression matrix and offers a model-free scheme. Firstly, MRAC is applied to adjust the control gains, then TDE is implemented to estimate the unknown dynamical robotic system, and NN is employed to deal with the TDE estimation error. The overall stability of the robotic dynamics is investigated using the Lyapunov theorem. In the end, computer simulations are compared to validate the effectiveness of the proposed scheme.

With the unquestionable commercial success of air cargo transportation, cargo loading is a crucial step that selects the optimal placement solution for a given aircraft hold and a set of cargoes. This combinatorial optimization promotes airlines’ revenue (e.g., minimizing fuel consumption) with the encoded constraints in the solution space. In practical scenarios, cargo loading includes dozens of loading constraints (e.g., isolation of dangerous cargoes). However, existing techniques either over-simplify such constraints due to the expensive manual modeling in combinatorial optimization, or suffer from a time-consuming optimization process due to the large search space in heuristic search. In this paper, we present FastLoader, an optimization acceleration approach that employs large language models (LLMs) to distinguish critical structural patterns in the simulated cargo loading data while still scaling to numerous loading constraints in real scenarios. FastLoader’s key design features are the following: (i) a cargo loading constructor, which converts the information of both cargo types and loading constraints into pre-defined data structures, thus avoiding manual modeling and improving solution accuracy; (ii) a cargo loading solver and a search space reducer, which work together to effectively reduce search space and accelerate the optimization process. We evaluate the proposed approach using a list of practical scenarios from industry transportation systems, and the results show the followin: FastLoader improves accuracy by 10% compared to combinatorial optimization, and reduces the optimization time by 90% with 1.5% accuracy losses compared to heuristic search.

In this paper, we apply data-driven optimization to estimate key parameters in a metric graph-based epidemiological model, with the aim of analyzing the effect of road networks on the geographic spread of epidemics. As a case study, we fit our model to data from the COVID-19 pandemic in Poland during 2021. Our dataset integrates county-level daily case reports, national census information, and traffic flow studies. This framework allows us to examine the relative contribution of specific travel routes over time and infer unobserved transmission patterns in the presence of incomplete or unreliable case reporting. The optimization problem that arises from the model fitting yields an objective function resembling the Rosenbrock “banana” or “valley” function, a classical difficult benchmark for optimization algorithms. To our knowledge, this represents the first appearance of a Rosenbrock-type function in a real-world epidemiological context. We demonstrate that such a structure can emerge naturally from a simple uncoupled SIR model under specific conditions: a low initial incidence rate and a prolonged infectious period. This suggests that the Rosenbrock behavior is an intrinsic feature of fitting compartmental models to approximately Gaussian epidemiological data, providing a realistic yet simple scenario with which to test optimization algorithms. We explore optimization strategies suited to the Rosenbrock-type structure and identify a feasible parameter set for modeling the spread of COVID-19 in Poland. We use this set of parameters to identify discrepancies between the model and the data, explore how reducing traffic flow into urban areas can help flatten the infection curve, and identify some patterns in the distribution of intra-versus inter-city incidence rates. While recognizing the complex interplay of social and behavioral elements that cannot be fully captured in a high-level geographic model, our findings highlight the usefulness of metric graph-based models for understanding large-scale disease transmission in structured transportation networks.