Search Results (9,373)

Artificial-cilia-driven transport is an important phenomenon in microfluidic systems, biomedical devices, porous transport media, and thermal-management applications, where controlled momentum, heat, and mass transfer are required. Accurate prediction of such coupled transport processes becomes challenging when magnetic effects, porous resistance, variable thermophysical properties, and chemical reactions are considered simultaneously. In the present study, we focus on the numerical and machine-learning prediction of artificial-cilia-driven ethylene glycol–silver (Ag–EG) nanofluid flow through a curved porous channel. The mathematical model incorporates Darcy–Brinkman–Forchheimer resistance, magnetic effects, viscous dissipation, heat generation, variable viscosity, thermal conductivity and diffusivity, and Arrhenius activation-energy effects. The governing nonlinear momentum, energy, and concentration equations are formulated in curvilinear coordinates under low-Reynolds-number conditions and solved using an implicit finite-difference method. A feed-forward artificial neural network is developed using 5000 numerical samples, which are divided into 70% training, 15% validation, and 15% testing subsets. The Levenberg–Marquardt algorithm is employed for network training, while MSE, RMSE, and (R²) are used to evaluate prediction accuracy. The results show that increasing the Darcy number enhances fluid velocity because of greater porous-medium permeability, whereas increasing the magnetic and Forchheimer parameters reduces the flow through Lorentz-force and nonlinear porous resistance, respectively. Increasing the thermal-conductivity parameter from 0.0 to 0.6 decreases the maximum temperature by about 15%, while increasing the activation-energy parameter from 0.1 to 1.0 increases the average species concentration by approximately 18–22%. The ANN predictions show close agreement with the numerical results. The developed numerical–machine-learning framework provides an efficient approach for predicting coupled transport behavior in cilia-driven porous-channel systems. Full article

To address the coordinated control of rock-breaking energy efficiency and construction-induced vibration of a twin-wheel trench cutter in vibration-sensitive urban environments, a three-dimensional explicit dynamic model of twin-wheel rock cutting was established based on the JH-2 constitutive model, and multi-objective optimization of construction parameters was performed by combining response surface methodology (RSM) with the NSGA-II algorithm. The applicability of the adopted JH-2 parameter set for describing the dynamic response of sandstone was evaluated through comparison with split Hopkinson pressure bar (SHPB) test results. On this basis, the stress response, damage propagation, fragmentation evolution, and vibration characteristics of the rock under the successive action of multiple cutter teeth were investigated. The results show that rock failure within the twin-wheel interaction zone evolves successively from localized damage and damage-band propagation to coalescence of the damaged zones between the two cutter wheels and continuous fragmentation. Free surfaces generated by preceding fragmentation reduce the confinement of subsequent cutter–rock interaction zones and facilitate further rock breakage. The cutting load exhibits pronounced intermittent impact and non-stationary characteristics, while the induced rock vibration shows broadband and multi-frequency responses. The peak particle velocity (PPV) generally decreases with increasing propagation distance and is positively correlated with the milling force. Response surface models were established with specific energy (SE) and PPV as the optimization objectives. In terms of the linear effects, SE is influenced, in descending order, by milling thickness, cutting angle, and milling speed, whereas PPV is mainly governed by milling thickness and milling speed, with milling angle exhibiting a pronounced nonlinear effect. Increasing the milling thickness reduces SE but simultaneously increases PPV, demonstrating a distinct trade-off between rock-breaking energy efficiency and vibration control. The proposed SE–PPV multi-objective optimization framework provides a basis for selecting twin-wheel cutting parameters that balance efficient rock breaking and vibration control in vibration-sensitive urban environments. Full article

Due to coupling in industrial control structures, alarm variables may exhibit interconnected variations, leading to the “alarm-triggered alarm” phenomenon, which can contribute to alarm flooding. Causal inference methods can characterize dependencies among alarm variables and help identify potential propagation paths for alarm configuration optimization. However, strong autocorrelation, nonlinear dynamics, and time-delay effects in complex industrial processes may lead conventional causal inference methods to identify spurious causal relationships, thereby reducing the reliability of the resulting causal graphs. To address this issue, we propose a Liang–Kleeman information flow–Max-Min Hill-Climbing (IF-MMHC) causal graph framework that combines physics-informed information flow analysis with constrained score-based structure learning. Specifically, Liang–Kleeman information flow (LKIF) separates endogenous evolution from external driving effects and provides physically informed constraints for identifying candidate causal relationships, thereby reducing spurious causal effects associated with temporal dependence. The resulting candidate graph is then optimized using the Max-Min Hill-Climbing (MMHC) algorithm through constraint checking and score-based structural search, further removing redundant and indirect edges and improving the structural parsimony of the causal graph. Experiments on multiple fault scenarios of the Tennessee Eastman Process demonstrate the effectiveness of IF-MMHC in identifying key alarms and causal propagation relationships while reducing redundant causal connections. The results provide evidence for the applicability of the proposed framework to industrial alarm analysis and configuration optimization, while broader validation across different industrial processes remains necessary. Full article

Accurate extraction of photovoltaic (PV) model parameters is essential for reliable PV-system simulation, maximum power point tracking, fault diagnosis, and performance assessment. However, this task is a nonlinear, multimodal, and strongly coupled optimization problem, particularly for double-diode models. This paper proposes an Adaptive Oscillatory Flow Direction Algorithm (AOFDA) for accurate parameter estimation of single-diode and double-diode PV models. AOFDA extends the Flow Direction Algorithm with three mechanisms: adaptive oscillatory flow movement for population diversification and directional exploration, adaptive exploration–exploitation probability to regulate the global-to-local search transition, and an elite-guided learning strategy to enhance information exchange among high-quality solutions. A Lambert W-based fitness-evaluation scheme is employed to calculate the PV output current explicitly, avoiding repeated iterative solutions and reducing computational burden. The method is assessed on four PV datasets (RTC France, amorphous silicon, PWP201, and LSM20) and compared with the original FDA and several metaheuristics using RMSE, boxplots, and Wilcoxon tests. Results show that AOFDA provides accurate and stable PV parameter estimates, achieving RMSE values of 6.460534 × 10⁻⁴ for RTC France and 2.454593 × 10⁻⁵ for the amorphous-silicon double-diode model. Overall, AOFDA improves convergence consistency and estimation accuracy, especially for higher-dimensional PV parameter-identification problems. Full article

Complex optimization problems in engineering and data-driven modeling require effective exploration, stable convergence, and adaptability across different decision representations. This study proposes an Adaptive Reinforced Elite Gray Wolf Optimizer (AREGWO) that combines opposition-based initialization, nonlinear convergence control, reinforced elite guidance, adaptive weak perturbation, stagnation-aware partial restart, and greedy elite inheritance. AREGWO is evaluated through CEC2022 benchmark optimization, synthetic 2D constrained robotic path planning, and football pass-quality prediction. On CEC2022, AREGWO ranks second among nine algorithms in the Friedman test and significantly outperforms seven of eight competitors in the Wilcoxon signed-rank test, while L-SHADE remains the strongest baseline. Component-wise ablation shows that the six mechanisms provide complementary and problem-dependent contributions. In robotic path planning, 30 independent runs on six random maps yield a six-map average path length of 134.38 m, which is 3.61–5.12% shorter than those obtained by the four population-based baselines and 1.08% shorter than A*, with a 100% success rate in all tested environments. For football pass-quality prediction, AREGWO is coupled with TabM for outer-level hyperparameter optimization of a predefined pass-quality reference score. AREGWO-TabM achieves an RMSE of 0.445, an MSE of 0.198, and an R² of 0.920, providing the best RMSE, MSE, and R² among the evaluated neural models, although ResNet attains a lower MAE. These results demonstrate the competitive search capability of AREGWO across the numerical, synthetic spatial, and model-level optimization settings evaluated in this study. Full article

Microgrid scheduling is a nonlinear, multimodal, and highly constrained optimization problem that requires effective coordination among renewable generation, controllable distributed generators, energy storage, and grid interaction. To improve the search capability of the original Crayfish Optimization Algorithm (COA), this study proposes an Enhanced Crayfish Optimization Algorithm (ECOA) inspired by the coordination characteristics of cross-media performing arts. Rather than simply combining independent operators, ECOA constructs a state-aware and behavior-specific adaptive search framework around the intrinsic behavioral structure of COA. Specifically, a rhythm-guided adaptive search strategy is embedded into the summer-resort behavior to dynamically regulate search intensity, a multi-role collaborative guidance strategy reconstructs the competition behavior through time-varying information interaction, and a stagnation-aware stage-transition perturbation strategy selectively reactivates exploration when the search becomes stagnant. Extensive experiments are conducted on the CEC2014, CEC2017, and CEC2022 benchmark suites under six-dimensional settings, and the results are compared with those of nine representative metaheuristic algorithms. The results show that ECOA achieves highly competitive convergence accuracy and run-to-run stability on most benchmark functions. Wilcoxon rank-sum tests and Cliff’s delta effect-size analysis indicate that many of the observed improvements are statistically significant and practically meaningful, while the Friedman test ranks ECOA first in all six benchmark settings. Ablation experiments further demonstrate that all three mechanisms contribute positively to performance and that their coordinated integration is more effective than any individual strategy. Parameter-sensitivity analysis shows that ECOA maintains effective convergence over a relatively broad range of the disturbance coefficient λ, with λ = 0.5 providing a balanced setting across different landscapes. Finally, ECOA is applied to a deterministic day-ahead microgrid scheduling problem with explicit feasibility repair and power-balance penalty handling. Under the considered operating scenario, ECOA achieves the lowest mean operating cost of 1361.8028 among all compared algorithms and produces coordinated power-allocation schedules satisfying the modeled operational constraints. These results demonstrate the effectiveness of ECOA for complex numerical optimization and constrained engineering scheduling problems. Full article

Since different geographical origins of black beans have different nutritional and economic values, their classification is of great significance in the field of agriculture. Therefore, this study aims to develop a classification model for black bean origin identification based on chemometric analysis of near-infrared (NIR) spectral data. Due to the overlap and complex distribution characteristics of raw NIR spectral data, traditional linear discriminant analysis algorithms are unable to extract effective discriminative information from them. To overcome these limitations and improve the performance of the algorithms, fuzzy set theory was incorporated into orthogonal linear discriminant analysis (OLDA), and the fuzzy OLDA (FOLDA) was proposed to extract discriminant features from the NIR absorption spectra of black beans. FOLDA enhances sample diversity and better handles noisy data. Furthermore, by combining FOLDA with a linear kernel function, the kernel FOLDA (KFOLDA) algorithm was proposed to address the challenge of linear algorithms in processing nonlinear data. KFOLDA not only effectively captures and processes nonlinear information within the data but also significantly improves the model’s classification accuracy. The experimental results show that the K-nearest neighbor (KNN) classifier achieved classification accuracies of 86.4% and 95.5% for FOLDA and KFOLDA, respectively. Thus, the proposed model demonstrates satisfactory classification accuracy and performance metrics for distinguishing different geographical origins of black beans. Full article

Growing global demand for resilient and sustainable infrastructure requires seismic retrofit strategies that simultaneously enhance structural performance and minimise environmental impact. This study presents a multi-objective optimisation framework that integrates circular economy (CE) indicators with seismic performance evaluation for the retrofit design of reinforced concrete (RC) shear walls. Nonlinear finite element simulations were carried out in SeismoStruct to evaluate a 125 mm thick component-level RC shear wall strengthened using three techniques: carbon-fibre-reinforced polymer (CFRP) overlays, externally bonded steel plates, and RC jacketing. Structural performance was assessed through nonlinear pushover and cyclic analyses interpreted using the Eurocode 8 N2 method and the ASCE 41–23 displacement coefficient method, while environmental and circularity performance was quantified using the Material Circularity Index (MCI), the Carbon Equivalence Damage Index (CEDI), and a Design for Circularity (DfC) score. A Non-dominated Sorting Genetic Algorithm II (NSGA-II) was used to generate Pareto-optimal retrofit solutions across four objectives: seismic capacity, embodied carbon, retrofit cost, and material circularity. Results show lateral-capacity increases of approximately 50–75% for steel strengthening and RC jacketing, while CFRP overlays leave the flexural capacity unchanged but double the shear capacity: steel plate strengthening provides the highest structural capacity and circularity performance ($\mathrm{MCI}=0.45$, $\mathrm{DfC}=8.7/10$), RC jacketing yields the greatest stiffness gain but the lowest circularity ($\mathrm{MCI}=0.08$), and CFRP overlays offer the most balanced structural, environmental, and economic performance. The framework was validated on a six-storey RC building containing 200 mm thick shear walls, confirming that integrating seismic assessment with circular economy indicators supports more sustainable and resilient retrofit decision-making. Full article

The safety of lithium-ion batteries is a critical concern, and this paper introduces a novel simulation approach for thermal runaway using the control volume free element method (CVFrEM). This method uniquely integrates the geometric adaptability of mesh-free techniques with the local conservation properties inherent to the finite volume method. The governing equations for heat generation and heat transfer within the battery are discretized through the CVFrEM framework, thereby ensuring strict energy conservation at the local control volume level. To accurately capture the nonlinear Arrhenius kinetics governing the internal reactions, a coupled solution algorithm is implemented within the CVFrEM formulation. The proposed method is rigorously validated against both experimental data and reference results, confirming its predictive accuracy and exceptional numerical stability, particularly in scenarios involving steep thermal gradients. Overall, this work offers a promising numerical tool for predicting thermal runaway events and provides valuable insights for advancing battery safety simulations. Full article

This study investigates a dynamic parallel-machine scheduling problem with human–robot collaboration (DPMS-HRC), where dynamically arriving jobs are assigned to eligible workstations and processed by workers, robots, or human–robot collaboration based on real-time system states. Worker-involving modes improve operational flexibility but may increase actual processing times and cause uneven fatigue distribution due to fatigue accumulation. To coordinate production responsiveness and fatigue management, a weighted objective combining average flow time and the worker fatigue balance index is minimised. A mixed-integer nonlinear programming (MINLP) formulation is developed to model dynamic job arrivals, alternative processing modes, worker–robot allocation, and fatigue-dependent processing times. An action-masked and potential-shaped multi-agent proximal policy optimisation (AMPS-MAPPO) algorithm is then designed. Action masking restricts policy sampling to structurally feasible job-mode-resource combinations, while time-scaled potential shaping provides dense feedback for long-horizon learning. Experiments on dynamic instances of different scales show that AMPS-MAPPO achieves millisecond-level online inference, reduces average flow time by 40.98% relative to the best dispatching-rule benchmark and by 8.17–20.92% relative to deep reinforcement learning baselines, and obtains the lowest overall worker fatigue balance index among the compared deep reinforcement learning algorithms. Full article

In wireless sensor networks (WSNs), reliable link-quality estimation depends strongly on channel characteristics, which can be flexibly captured by the $\kappa$-$\mu$ shadowed fading model in complex closed metallic cavity (CMC) environments. However, a fundamental tension arises: conventional numerical methods for estimating the parameters of this model generally rely on the iterative solution of coupled nonlinear moment equations, which may introduce considerable computational cost and sensitivity to initialization and convergence settings, thereby limiting their applicability to rapid channel characterization in WSNs. We propose a closed-form multi-parameter estimation algorithm based on the method of moments that computes parameters directly without recourse to iterative numerical solvers. The key insight is that restructuring the normalized moment equations reveals a hidden quadratic structure, enabling explicit parameter extraction without iterative solver convergence requirements. Simulation results demonstrate sub-millisecond closed-form evaluation in the representative MATLAB implementation and substantial computational savings over the considered iterative solvers. Under the same prescribed composite parameter, it achieves essentially the same distribution-fitting accuracy as the corresponding constrained iterative estimator while requiring substantially less computation time. Furthermore, the reconstructed fading distributions provide accurate ergodic-capacity predictions, indicating the potential of the proposed method for computationally efficient link-quality characterization in WSNs. Full article

To address the limitations of conventional constant-rate pesticide application in adapting to spatial variations in crop canopy structure, this study proposes a pesticide spraying strategy optimization method for a tracked air-assisted spraying robot based on a novel Virginia Creeper Optimization Algorithm (VCOA). First, a multivariable nonlinear constrained optimization model is established by considering the coupled effects of robot chassis travelling speed, ducted-fan rotational speed, pump pressure, equivalent nozzle atomization diameter parameter and spray-boom pitch angle. A comprehensive spraying effectiveness objective function is constructed and maximized objective function is constructed to coordinate droplet deposition performance, canopy coverage uniformity, pesticide utilization efficiency, off-target loss, and operational energy consumption. Then, inspired by the random spreading, adhesive climbing, and phototropic growth mechanisms of Virginia creeper, VCOA is developed to enhance the balance between global exploration and local exploitation in complex non-convex optimization problems. The convergence behavior and optimization capability of the proposed algorithm are evaluated using benchmark functions and comparative algorithms. Furthermore, the algorithm is introduced into the variable-rate spraying control process of a tracked air-assisted spraying robot, and its practical effectiveness is verified through simulation and field spraying experiments. The results show that VCOA exhibits superior convergence robustness and global search capability compared with the selected benchmark algorithms. In field experiments, the VCOA-based spraying strategy significantly improves droplet deposition coverage, deposition density, and spatial uniformity, demonstrating its effectiveness and application potential for precision pesticide application in intelligent plant protection equipment. Full article

The traditional control approach for piston-type wave makers is reduced to a piston position tracking problem based on open-loop wave-maker theory; however, due to hydrodynamic loading, nonlinear effects, and propagation delays, the target wave condition at the measurement section cannot be guaranteed. This experimental study involves controllers tuned offline through numerical simulations and validated in real time on a 24 m wave channel. The novelty of this work is a two-layer field-programmable gate array (FPGA) cascade that regulates the measured wave height in closed loop, not just the piston position. In the inner layer, the hydraulic servo cylinder position is controlled within a 1 ms cycle using a proportional–integral (PI) or a Takagi–Sugeno (T–S) fuzzy controller; the outer layer, in a 20 ms cycle, updates the position reference via real-time wave height feedback. Controller parameters were tuned using three meta-heuristic algorithms under the integral of time-weighted absolute error criterion, and a similarity-based rule reduction reduced the T–S consequent parameters from 27 to 9, yielding 66.7% and 62.5% reductions in FPGA execution time and digital signal processing slice usage with tracking-error changes of only 2.2–2.3%. Experiments confirm high-accuracy position tracking and steady-state wave height regulation, verified by fast Fourier transform analysis. Full article

Metaheuristic optimization algorithms have become one of the most important research areas for solving complex, nonlinear, and high-dimensional optimization problems. These algorithms are widely used in many application domains such as engineering, artificial intelligence, energy systems, healthcare, communication technologies, manufacturing, data science, and industrial applications. Their strong exploration and exploitation capabilities enable them to provide effective solutions for optimization problems where classical optimization methods are insufficient. In addition, many metaheuristic optimization algorithms attempt to establish a balanced and adaptive search behavior between exploration and exploitation processes, reflecting an important form of search symmetry during the optimization process. This systematic review presents a comprehensive analysis of metaheuristic optimization algorithms based on 515 distinct metaheuristic optimization algorithms. In this study, the algorithms are systematically classified into eleven major categories according to their inspiration sources and dominant search behaviors: Evolutionary-Based, Swarm Intelligence-Based, Bio-Inspired, Ecological/Environmental-Based, Physics-Based, Mathematics-Based, Human-Based, Game/Sport-Based, Chemical and Molecular-Based, Music-, Art-, and Culture-Inspired, and Hybrid and Advanced Metaheuristics. The review analyzes the historical evolution, inspiration diversity, search structures, symmetry-oriented search balance, and development trends of modern metaheuristic optimization algorithms. This study also provides a broader understanding of how metaheuristic optimization algorithms have evolved structurally and conceptually over time. In addition, the study highlights the importance of balanced and adaptive search mechanisms for maintaining optimization efficiency under complex problem conditions. The findings show that the number and diversity of metaheuristic optimization algorithms have increased significantly, especially after 2015. Population-based search structures remain the dominant optimization approach in the literature, while hybrid optimization methods, adaptive search mechanisms, and artificial intelligence integration have become major research trends in recent years. The study also discusses important challenges in the field, including algorithm similarity, benchmark standardization, theoretical limitations, and fair performance evaluation. Full article

Logistics network development and supply chain organizational innovation may create external conditions for corporate green innovation. Using 30,700 firm-year observations for Chinese A-share listed companies from 2012 to 2024, this study examines the relationship between joint coverage by the Commercial Logistics Node City Program and the Supply Chain Innovation and Application Pilot Program and corporate green innovation. Double machine learning is employed to account flexibly for nonlinear relationships between policy coverage, firm characteristics, and green innovation. Dual-policy coverage is positively and significantly associated with green invention patent applications, with a baseline coefficient of 0.1131. The result remains positive under alternative cross-fitting folds, machine learning algorithms, sample restrictions, and measures of green innovation. Channel analyses indicate that dual-policy coverage is associated with greater supply chain digitalization, lower supply chain risk, and higher innovation efficiency. The estimated association is stronger among firms with higher R&D intensity and weaker among firms facing greater environmental uncertainty. These findings provide firm-level evidence that logistics connectivity and supply chain organizational conditions are relevant to corporate green innovation and offer policy implications for sustainable industrial development. Full article