Search Results (2,109)

To accurately characterize the complex coupling and nonlinear interactions between meteorological and oceanic variables in offshore wind power scenarios, this study proposes a novel forecasting model based on a “multi-scale fusion-decomposition-reconstruction-optimization-prediction” framework. This model integrates Variational Modal Decomposition (VMD) with the feature-interaction Informer framework, employing an enhanced Honey Badger Algorithm (HBA) for the collaborative optimization of their key parameters. The enhanced HBA integrates cubic chaotic mapping, random perturbation strategy, elite tangent search, and differential mutation operations to strengthen its global optimization capability and convergence efficiency. The model construction process proceeds as follows: First, sample entropy (SE) is applied to evaluate the entropy values and reconstruct sequences of the modal components obtained from VMD. Subsequently, the dynamic adjustment of the maximum information coefficient (DE-MIC) is employed to select key input variables from multi-source features. Subsequently, the feature interaction-probabilistic sparse attention mechanism (FI-ProbSparse-AM) unique to the feature interaction-based Informer is employed to construct an attention architecture capable of explicitly modeling dependencies among multidimensional variables, thereby effectively capturing the spatiotemporal latent correlations between wind power output and multi-source features. Experiments based on real offshore wind farm data demonstrate that the MAPE values are reduced by approximately 11% compared to existing benchmark models. The proposed method demonstrates significant advantages in both prediction accuracy and stability. Full article

The integration of inverter-based generation units, such as photovoltaic systems, wind turbines, and vehicle-to-grid (V2G) technologies, has introduced new challenges in maintaining power and frequency stability in modern power systems. Virtual Synchronous Generators (VSGs) have emerged as a promising solution to enhance system stability; however, existing control methods often lack the robustness and flexibility needed to address deliberate and unplanned outages effectively. This paper presents a novel approach for optimizing power control in generation units using a Long Short-Term Memory (LSTM)-based machine learning method. The proposed LSTM-based controller provides a fast and real-time response, ensuring robust and flexible performance under varying operational conditions. Unlike traditional controllers, the proposed method effectively handles nonlinearities and uncertainties associated with inverter-based units. Additionally, it effectively balances technical and economic aspects of power system operation by minimizing oscillations and optimizing resource utilization. The proposed approach is benchmarked against conventional control methods through a detailed simulation-based comparative analysis against a linear Model Predictive Control strategy under identical operating conditions. Simulation results indicate that the proposed controller reduces frequency deviations by up to 66.7%, voltage deviations by 62.5%, and total operational cost by approximately 11.3%, while achieving nearly 90% faster dynamic response, validating its effectiveness for modern power systems. Full article

The (2+1)-dimensional Boussinesq equation is a fundamental model in nonlinear wave theory, governing shallow-water wave propagation, coastal dynamics in ocean engineering, and long waves in geophysical fluid systems such as atmospheric and oceanic currents. We present a novel neural network symbolic computation framework that seamlessly integrates neural architectures for powerful function approximation with symbolic manipulation for exact algebraic resolution, eliminating the need for bilinear transformations and thereby substantially reducing computational complexity. Applying this framework, we derive five previously unreported exact analytical solutions using carefully designed neural network configurations and probe functions. These solutions provide valuable tools for modeling ocean internal waves, coastal engineering simulations, and nonlinear optical pulse dynamics. In practice, the method delivers faster and more accurate simulations, improving engineering design and environmental prediction capabilities. By synergistically combining neural networks with symbolic computation, our approach surpasses traditional numerical methods and physics-informed neural networks in both accuracy and efficiency, opening new avenues for solving complex nonlinear partial differential equations. Full article

The Kuramoto–Sivashinsky (KS) equation and its fractional form (FKS) are widely used across scientific fields, including fluid dynamics, plasma physics, and chemical processes, to model nonlinear phenomena such as shock waves. It is worth emphasizing that this contribution is part (II) of a larger, systematic research program aimed at modeling, for the first time, completely nonintegrable, nonplanar, and fractional nonplanar evolutionary wave equations. This work focuses on the nonplanar KS framework and its applications to dust–acoustic shock waves in a complex plasma composed of inertial dust grains and inertialess nonextensive ions. This study analyzes both the nonplanar integer KS and nonplanar FKS equations, accounting for geometric effects. This is because the nonplanar model is most suitable for analyzing various nonlinear phenomena (e.g., shock waves) that arise and propagate in plasma physics, fluids, and other physical and engineering systems. Since the nonplanar KS equation is a fully non-integrable problem, its analysis poses a significant challenge for studying the properties of nonplanar shock waves in plasma physics. Therefore, the primary objective of this study is to analyze the nonplanar KS equation using the Ansatz method, thereby deriving semi-analytical solutions that simulate the propagation mechanism of nonplanar shock waves in various physical systems. Following this, we investigate the effect of the fractional factor on the profiles of nonplanar dust–acoustic shock waves to elucidate their propagation mechanism and assess the impact of the memory factor on their behavior. To achieve the second goal, we face a significant challenge because the model under study does not support exact solutions and is more complex than simpler physical models. Thus, the Tantawy technique is employed to overcome this challenge and to analyze this model for generating highly accurate analytical approximations suitable for modeling nonplanar fractional shock waves in various plasma models and in other physical and engineering systems. Full article

In this research, the nonlinear bending behavior of sandwich beams with auxetic re-entrant cores and carbon nanotube-reinforced (CNT) face sheets are investigated using the von Kármán strain theory and the Generalized Differential Quadrature (GDQ) method. The results demonstrate the high accuracy of the GDQ method in solving nonlinear problems with a minimal number of grid points. Validation is performed by comparing the obtained results with those reported in previous studies. The findings indicate that CNT-reinforced composite beams exhibit superior bending performance compared to sandwich beams with re-entrant cores and conventional composite face sheets. Furthermore, a parametric study on the core geometry reveals that optimal bending performance is achieved when the interior angle (θ) of the core is approximately 0 to 2 degrees, and the nondimensional auxetic parameter η_1 is minimized. This study highlights the significance of nonlinear analysis, particularly for long and slender beams. Full article

This paper investigates a fairness-aware task assignment problem, where traditional models prioritize operational effectiveness while overlooking workload fairness. Motivated by real-world airport operations, we propose a multi-objective task assignment model that penalizes deviations between actual and expected workloads. Expected workloads are computed from the aggregated task density over each shift’s time window. To capture the nonlinear nature of perceived unfairness, we define the imbalance cost using a quadratic penalty function. We develop three piecewise linear approximations to solve the model efficiently and further simplify them by exploiting convexity to remove binary variables. Experiments on real-world data from a major airport in China show that the proposed methods significantly reduce solution time while preserving solution quality. Under a one-hour time budget, the piecewise linear approximation models achieve up to 5.12% cost savings in large-scale instances compared to the original nonlinear model. Moreover, our proposed fairness-aware task assignment model yields substantial improvements in workload balance (over 90%) at a limited cost to service quality (approximately 20%), even with small imbalance penalties. Full article

The nonlinear static (pushover) procedure for estimating target displacement is incorporated in Annex B of Eurocode 8 and is commonly implemented through the N2 method, originally developed at the University of Ljubljana. While advanced nonlinear dynamic analyses can provide detailed insight into seismic structural behavior, they require sophisticated modeling and significant computational effort. In contrast, simplified procedures are often preferred in engineering practice for preliminary assessment and design verification. This study evaluates the applicability of the N2 method by comparing analytically obtained target displacements with experimentally measured responses of a three-story model building subjected to seismic excitation. The nonlinear capacity curve derived from pushover analysis is transformed into the acceleration–displacement response spectrum (ADRS) domain and evaluated against the demand spectrum to determine the performance point. The experimentally obtained value, reported in a reference study, is used as a benchmark for assessment. The results indicate that the N2 method provides an approximate but rapid estimation of target displacement. The findings highlight both the practical usefulness and the inherent limitations of the N2 approach when applied to structures. Full article

Proton exchange membrane fuel cells (PEMFCs) are attractive energy sources for clean and efficient power generation; however, their nonlinear characteristics and sensitivity to operating condition variations make maximum power point tracking (MPPT) a challenging control problem. Conventional MPPT techniques often exhibit slow convergence, steady-state oscillations, and degraded performance under dynamic fuel flow variations. This paper proposes a machine learning–driven MPPT control strategy for a PEMFC system integrated with a DC–DC boost converter. The MPPT problem is formulated as a supervised classification task, where machine learning classifiers generate duty-cycle commands to regulate the converter and ensure operation at the maximum power point. A detailed PEMFC–converter model is developed in MATLAB/Simulink-2025b, and a dataset of 3000 labeled samples is generated under varying fuel flow conditions. Several classification algorithms, including decision trees, support vector machines (SVM), k-nearest neighbors (kNN), and ensemble learning methods, are systematically evaluated within an identical simulation framework. Simulation results show that the proposed machine learning-based MPPT controller significantly improves dynamic and steady-state performance. Ensemble Boosted Trees achieve the best overall response with a settling time of approximately 32 ms, peak power overshoot below 4.5%, and steady-state power ripple limited to 1.5%. Quadratic SVM and weighted kNN classifiers also demonstrate stable tracking behavior with power ripple below 2.1%, while overly complex models such as Cubic SVM suffer from large oscillations and reduced accuracy. These results confirm that classification-based machine learning offers an effective, fast, and robust MPPT solution for PEMFC systems under dynamic operating conditions. Full article

Backpropagation Neural Networks (BPNNs) are widely used in fault diagnosis and parameter prediction due to their simple structure and strong universal approximation capabilities. However, BPNNs suffer from slow convergence and susceptibility to poor local minima under basic gradient descent settings. To address these issues, this paper proposes a Balanced Grey Wolf Optimizer (BGWO) as an alternative to gradient descent for training BPNNs. This paper proposes a novel stochastic position update formula and a novel nonlinear convergence factor to balance the local exploitation and global exploration of the traditional Grey Wolf Optimizer. After exploration, the optimal convergence coefficient is determined. The test results on the six benchmark functions demonstrate that BGWO achieves better objective function values under fixed iteration settings. Based on BGWO, this paper constructs a training method for BPNN. Finally, three public datasets are used to test the BPNN trained with BGWO (BGWO-BPNN), the BPNN trained with Levenberg–Marquardt, and the traditional BPNN. The relative error and mean absolute percentage error of BPNNs’ prediction results are used for comparison. The Wilcoxon test is also performed. The test results show that, under the experimental settings of this paper, BGWO-BPNN achieves superior predictive performance. This demonstrates certain advantages of BGWO-BPNN. Full article

Accurate identification of road roughness is pivotal for optimizing vehicle suspension control and enhancing passenger comfort. However, existing data-driven methods often struggle to balance classification accuracy with the strict computational constraints of real-time onboard monitoring. To address this challenge, this paper proposes a lightweight and robust road roughness classification framework utilizing a single sprung mass accelerometer. First, to overcome the scarcity of labeled real-world data and the limitations of linear models, a high-fidelity co-simulation platform combining CarSim and Simulink is established. This platform generates physically consistent vibration datasets covering ISO A–F roughness levels, effectively capturing nonlinear suspension dynamics. Second, we introduce DB-MLP, a novel Dual-Branch Multi-Layer Perceptron architecture. In contrast to computationally intensive Transformer or RNN-based models, DB-MLP employs a dual-branch strategy with multi-resolution temporal projection to efficiently capture multi-scale dependencies, and integrates dual-domain (time and position-wise) feature transformation blocks for robust feature extraction. Experimental results demonstrate that DB-MLP achieves a superior accuracy of 98.5% with only 0.58 million parameters. Compared to leading baselines such as TimeMixer and InceptionTime, our model reduces inference latency by approximately 20 times (0.007 ms/sample) while maintaining competitive performance on the specific road classification task. This study provides a cost-effective, high-precision solution suitable for real-time deployment on embedded vehicle systems. Full article

With the advancement of the dual carbon goals and the rapid increase in the proportion of new energy installations, the power system faces multiple challenges including insufficient flexibility resources, intensified fluctuations in generation and load, and reduced operational safety. Integrated energy systems (IESs), serving as key platforms for integrating diverse energy sources and flexible resources, possess complex internal structures and limited individual regulation capabilities, making direct participation in grid dispatch and market interactions challenging. To achieve large-scale resource coordination and efficient utilization, this paper investigates external characteristic modeling and cluster aggregation optimization methods for IES, proposing a comprehensive technical framework spanning from individual external characteristic identification to cluster-level coordinated control. First, addressing the challenge of unified dispatch for heterogeneous resources within IES, this study proposes an external characteristic modeling method based on operational feasible region projection. It constructs models for the active power output boundary, marginal cost characteristics, and ramping rate of virtual power plants (VPPs), enabling quantitative representation of their overall regulation potential. Second, a cluster aggregation optimization model for integrated energy systems is established, incorporating regional autonomy. This model pursues multiple objectives: cost–benefit matching, maximizing renewable energy absorption rates, and minimizing peak external power purchases. The Gini coefficient and Shapley value method are introduced to ensure fairness and participation willingness among cluster members. Furthermore, an optimization mechanism incorporating key constraints such as cluster scale, grid interaction, and regulation complementarity is designed. The NSGA-II multi-objective genetic algorithm is employed to efficiently solve this high-dimensional nonlinear problem. Finally, simulation validation is conducted on a typical regional energy scenario based on the IEEE-57 node system. Results demonstrate that the proposed method achieves average daily cost savings of approximately 3955 CNY under the optimal aggregation scheme, reduces wind and solar curtailment rates to 5.38%, controls peak external power purchases within 2292 kW, and effectively incentivizes all entities to participate in coordinated regulation through a rational benefit distribution mechanism. Full article

The Kuramoto–Sivashinsky (KS) equation and its fractional generalizations (FKSs) arise as canonical models for a wide class of nonlinear dissipative–dispersive systems, including thin-film flows, combustion fronts, drift–wave turbulence in plasmas, and chemically reacting media, where shock-like and strongly localized structures play a central role in the dynamics. Despite their apparent simplicity, KS-type models become analytically intractable once higher-order dissipation, geometric effects, and memory (fractional) operators are incorporated, and standard perturbative or transform-based schemes often lead to cumbersome recursive structures, slow convergence, or severe restrictions on the initial data. In this work, a novel direct approximation procedure, referred to as the Tantawy Technique (TT), is developed and implemented to solve and analyze planar fractional KS-type equations and their Burgers-type reductions in a systematic manner. The central difficulty is to construct, for a given physically motivated initial profile, a rapidly convergent series in fractional time that remains stable for a broad range of the fractional order and transport coefficients, while still retaining a clear link to the underlying shock-wave physics. To overcome this, the TT combines (i) a Tanh-based exact shock solution of the planar integer-order KS equation, obtained first as a reference via the standard Tanh method, with (ii) a carefully designed fractional-time ansatz in powers of $t^{\rho }$, where the spatial coefficients are determined recursively from the governing equation in the Caputo sense. This construction yields closed-form expressions for the first few terms in the approximation hierarchy and allows one to monitor convergence through residual and absolute error measures. Full article

The exponential growth of digital data and the escalating sophistication of cyber threats have intensified the demand for secure yet computationally efficient encryption methods. Conventional algorithms (e.g., AES-based schemes) are cryptographically strong and widely deployed; however, some implementations can face performance bottlenecks in large-scale or real-time workloads. While many modern systems seed from hardware entropy sources and employ standardized cryptographic PRNGs/DRBGs, security can still be degraded in practice by weak entropy initialization, misconfiguration, or the use of non-cryptographic deterministic generators in certain environments. To address these gaps, this study introduces FileCipher. This novel file-encryption framework integrates a chaos-enhanced Cryptographically Secure Pseudorandom Number Generator (CPRNG) based on the State-Based Tent Map (SBTM). The proposed design achieves a balanced trade-off between security and efficiency through dynamic key generation, adaptive block reshaping, and structured confusion–diffusion processes. The SBTM-driven CPRNG introduces adaptive seeding and multi-key feedback, ensuring high entropy and sensitivity to initial conditions. A multi-threaded Java implementation demonstrates approximately 60% reduction in encryption time compared with AES-CBC, validating FileCipher’s scalability in parallel execution environments. Statistical evaluations using NIST SP 800-22, SP 800-90B, Dieharder, and TestU01 confirm superior randomness with over 99% pass rates, while Avalanche Effect analysis indicates bit-change ratios near 50%, proving strong diffusion characteristics. The results highlight FileCipher’s novelty in combining nonlinear chaotic dynamics with lightweight parallel architecture, offering a robust, platform-independent solution for secure data storage and transmission. Ultimately, this paper contributes a reproducible, entropy-stable, and high-performance cryptographic mechanism that redefines the efficiency–security balance in modern encryption systems. Full article

With the continuous increase in high-speed train operating speeds, effective vibration suppression of the car body is critical for ensuring passenger comfort. This study proposes a composite damping device based on particle damping technology, featuring a variable cavity structure incorporating spring components designed for space-constrained areas. The primary aim of this work is to elucidate the energy dissipation mechanism of granular media under adaptive boundary conditions and to establish a novel method for overcoming the saturation limitations of traditional fixed-cavity dampers. The energy dissipation characteristics were investigated using coupled Discrete Element Method (DEM) and Multibody Dynamics (MBD) numerical simulations. Parametric analysis quantitatively demonstrated significant performance variations: 2 mm particles outperformed larger diameters by maximizing collision frequency, and cast iron particles (29.497 J) achieved approximately five times the energy dissipation of steel particles (5.909 J). Furthermore, the filling rate exhibited a non-linear relationship with damping performance, peaking at a 98% filling rate (57.251 J)—a nearly 9-fold increase compared to a 90% filling rate. Most notably, quantitative comparison confirms that the introduction of the spring-adaptive mechanism enhanced the total energy dissipation to approximately 2 times that of the traditional fixed-cavity design. Simulation results reveal that the flexible cavity significantly enhances performance by preventing particle packing and stagnation. The dynamic deformation continuously “recruits” particles into high-energy collision regimes, ensuring sustained broadband attenuation. These findings establish the spring-based variable volume design as a high-efficiency strategy for high-speed rail applications. Full article

We develop a novel Steffensen-type iterative solver to solve nonlinear scalar equations without requiring derivatives. A two-parameter one-step scheme without memory is first introduced and analyzed. Its optimal quadratic convergence is then established. To enhance the convergence rate without additional functional evaluations, we extend the scheme by incorporating memory through adaptively updated accelerator parameters. These parameters are approximated by Newton interpolation polynomials constructed from previously computed values, yielding a derivative-free method with R-rate of convergence of approximately 3.56155. A dynamical system analysis based on attraction basins demonstrates enlarged convergence regions compared to Steffensen-type methods without memory. Numerical experiments further confirm the accuracy of the proposed scheme for solving nonlinear equations. Full article