Search Results (272)

Harmful algal blooms (HABs) caused by Karenia brevis (K. brevis) present a persistent ecological and public health challenge across coastal Florida. Reliable bloom forecasting is critical for protecting public health, supporting coastal economies, and enabling timely management responses. This study develops a regionally integrated machine learning framework to predict weekly K. brevis bloom occurrence using environmental data from both the Peace and Caloosahatchee Rivers, combined with coastal bloom records from Southwest Florida and Tampa Bay to enhance the spatial and temporal continuity of the response record. A Random Forest classifier was trained on a multi-decadal dataset incorporating river discharge, nutrient concentrations (total nitrogen and total phosphorus), wind forcing, sea surface temperature, salinity, and sea surface height anomalies as a proxy for Loop Current variability. The model achieved strong predictive performance on a chronologically withheld test set, with an overall accuracy of ~90%, balanced accuracy of 87.6%, and ROC–AUC of 0.972, indicating strong discrimination between bloom and non-bloom conditions with high precision and recall for bloom events. Bloom timing and persistence were captured with strong agreement during ongoing bloom periods, while non-bloom conditions were identified with low false-positive rates. Feature-response analyses indicated that bloom probability increased most sharply under moderate discharge and nutrient conditions, with diminished sensitivity at higher extremes. Learning curve analysis demonstrated robust training performance and stable generalization, with validation accuracy plateauing near 84%, suggesting a data-limited ceiling on forecast skill. By aggregating nutrient inputs across multiple watersheds and integrating spatially aligned bloom observations, this study demonstrates the utility of multi-source machine learning frameworks for regional-scale HAB prediction. The results support the development of early warning tools and provide a reproducible foundation for evaluating how combined watershed loading and physical forcing are associated with K. brevis bloom occurrence in complex estuary systems with watershed and coastal coupling. Full article

To address the limited accuracy of power demand prediction for hybrid electric vehicles under complex and dynamic driving conditions, this paper proposes a hybrid prediction approach based on the cascade correction of Long Short-Term Memory networks and Least Squares Support Vector Machines (LSTM-LSSVM-CC). The proposed method adopts a stage-wise modeling framework that exploits the least-squares optimality of LSSVM for low-frequency steady-state signals and the dynamic compensation capability of LSTM for high-frequency non-stationary residuals, thereby achieving complementary feature representation in the frequency domain. Specifically, an LSSVM is first used to construct a baseline regression model that captures stationary components, followed by an LSTM network that performs deep temporal modeling of the residual sequence to correct nonlinear prediction errors. Extensive experiments conducted on three standard driving cycles—CLTC-P, WLTP, and UDDS—demonstrate that the proposed model consistently outperforms conventional methods including LSSVM, RNN, ELMAN, and Random Forest in multi-step predictions, achieving an average RMSE reduction of 28–52% and maintaining correlation coefficients (R²) between 0.87 and 0.99. Particularly under highly dynamic and abrupt load conditions, the model exhibits superior real-time performance and stability while significantly mitigating cumulative prediction errors. These results demonstrate that the proposed LSTM-LSSVM-CC model achieves robust modeling performance of non-stationary time series while balancing prediction accuracy and computational efficiency, providing an effective technical foundation for hybrid vehicle energy management optimization and offering a transferable theoretical framework for time-series prediction in complex systems. Full article

Ground reaction forces (GRFs) are essential for assessing running biomechanics, and the combination of wearable sensors and machine learning offers an accessible alternative for estimating GRFs outside controlled environments. This scoping review summarized current methods used to predict GRFs during running. A structured search (2019–2025) identified 36 studies, from which 37% did not report participant’s training status, and 59% of all participants were males. Treadmill running was assessed in 58% of studies, which included larger samples (median N = 28) and more steps/participant (median = 65) than overground studies (median N = 14; median = 32). Deep learning models, particularly LSTM and Bi-LSTM networks, were the most applied techniques, though presenting similar accuracies compared to classical regression methods. Vertical GRF predictions were the most accurate, while mediolateral GRF predictions remain challenging. GRF-derived variables such as peak forces, impact peaks, and impulses were predicted more accurately than region-dependent metrics like loading rates. Notably, no study validated treadmill-trained models on overground running, limiting real-world generalizability. Future work should prioritize larger, sex-balanced cohorts, improving prediction of mediolateral GRFs and loading rates, and explore validating treadmill-based models in overground conditions. In conclusion, although machine learning shows promise for GRF predictions, key methodological gaps must be addressed to enable robust, real-world applications. Full article

The development of high-performance composite supercapacitors remains challenging because the specific capacitance of composite electrodes is jointly governed by electronic percolation, ion accessibility, and interfacial contact, all of which are strongly affected by the balance among active materials, conductive agents, and binders. Traditional equivalent circuit modeling and empirical trial-and-error methods are often inadequate for describing these non-linear relationships and optimizing electrode design. To address this limitation, we establish a physics-guided and interpretable machine learning (ML) framework for predicting the specific capacitance of composite electrodes. Unlike traditional methods that rely on macroscopic mass fractions, our approach constructs a feature space comprising ten descriptors, including two newly introduced binder-related proxy descriptors—Binder-to-Conductive Ratio (BCR) and Specific Binder Loading (SBL)—to better represent the influence of binder content. By systematically evaluating 17 machine learning algorithms on a high-fidelity dataset, we identify the XGBoost model, optimized via Bayesian optimization, as the best predictor, achieving a coefficient of determination (R²) of 0.981 and a low mean absolute percentage error (MAPE) of 14.49%. Importantly, interpretability analysis using Shapley Additive Explanations (SHAP) provides physically interpretable statistical insights, revealing that high BCR suppresses specific capacitance through an insulating barrier effect, whereas lattice distortion in the filler material promotes ion transport. This study offers a robust, data-driven framework for optimizing composite electrode performance, demonstrating the potential of interpretable ML models for the rational design of advanced energy-storage materials. Full article

The increase in utility-scale PV generation and the displacement of synchronous machines reduce system strength, reactive power headroom, voltage resilience, and overall power system stability, motivating a robust comparison of various mitigation technologies beyond static load-flow or PV assessments. RMS time-domain simulations are performed for balanced and unbalanced contingencies, and performance is quantified using post-fault voltage dip depth, undervoltage area (V < 0.9 pu.), recovery time to nominal, and RoCoF. These metrics are aggregated into a single weighted composite severity score, which is then normalised to the baseline to form the dynamic voltage resilience index (DVRI) and the Frequency Disturbance Relative Index (FDRI). The results show that the converter-based reactive power support devices deliver the fastest and most controllable post-fault voltage restoration, with the STATCOM achieving the lowest composite penalty and best DVRI under severe fault conditions but the poorest FDRI during PV plant trip/reconnection events. The synchronous condenser (SC) improves post-fault recovery through excitation driven reactive capability and increased short-circuit contribution, but its recovery to nominal voltage levels is slower and can produce negative-sequence current under unbalanced fault conditions whilst producing the smallest frequency disturbance and best FDRI. The SVC provides effective steady-state regulation but becomes less effective during extremely low voltages due to the voltage-dependent reactive power output, and its FDRI remains close to baseline. The BESS-GFM is dependent on the inverter current limits and the control priorities, which influence both voltage recovery and response times, achieving an FDRI scoring second to the SC. These metrics are combined into baseline normalised composite indices (DVRI and FDRI) using explicitly dimensionless sub-metrics (dip magnitude, exposure area, and recovery delay for voltage and deviation magnitude, windowed RoCoF, and exposure for frequency). Equal weights are used as a neutral baseline, and a weight sensitivity study is included to confirm that technology rankings are robust to plausible variations in weighting choice. Full article

Decarbonizing downstream steel logistics remains underexplored in sustainable supply chain management. This study proposes a machine learning-enhanced tri-objective optimization framework for the ship stowage planning problem (SSPP). The framework handles heterogeneous finished steel products, including coils, plates, ingots, tubes, and sections. The model simultaneously maximizes deadweight utilization and minimizes a novel Adaptive Weighted Moment Balance (AWMB) index. It also minimizes voyage carbon emissions through a trim-and-heel resistance penalty. A spatial-to-sequential discretization strategy transforms the NP-hard placement problem into a tractable permutation optimization. A deep neural network (DNN) surrogate achieves a 3.57-fold speedup with only 1.52% hypervolume degradation. An improved NSGA-III algorithm with adaptive operators ensures Pareto front exploration. Embedded step-wise moment verification guarantees dynamic stability throughout loading and unloading. Validated on real data from a Chinese steel enterprise, the framework achieves 99.88% deadweight utilization, reduces transverse and longitudinal imbalance by 48.27% and 90.54%, and cuts CO₂ emissions by 95.5% per voyage. SOLAS constraints, load line limits, and CII/FuelEU targets are addressed through embedded stability and capacity constraints. Multi-route and weather-dependent validation remains necessary before fleet-scale deployment. Full article

Objective biomarkers for neurodevelopmental disorders remain an unmet clinical need. The electroretinogram (ERG), a non-invasive recording of the retinal response to light, has shown promise as a physiological marker for autism spectrum disorder (ASD) and attention deficit/hyperactivity disorder (ADHD), yet existing classification approaches based on time-domain and time–frequency features achieve limited accuracy in clinically relevant multi-group scenarios. This study introduces ERG-Graph, a novel graph signal processing (GSP) framework that transforms each ERG waveform into a weighted, undirected graph through amplitude quantization and temporal-adjacency connectivity. Nine topological and spectral features, including total load centrality, clique number, algebraic connectivity, and clustering coefficient, were extracted from each graph to characterize the structural dynamics of the signal. Using light-adapted ERG recordings from 278 participants (ASD = 77, ADHD = 43, ASD + ADHD = 21, Control = 137), we evaluated these features across binary, three-group, and four-group classification scenarios using seven machine learning classifiers with 10-fold subject-wise cross-validation. The proposed ERG-Graph features achieved balanced accuracies of 0.91 (ASD vs. control, males) and 0.88 (ADHD vs. control, females). Critically, fusing ERG-Graph with time-domain features yielded a balanced accuracy of 0.81 for three-group classification (ASD vs. ADHD vs. control), representing an 11-percentage-point improvement over the previous benchmark of 0.70. Statistical analysis confirmed significant topological differences between groups (Kruskal–Wallis, p < 0.001; Cliff’s delta: large effect sizes), and SHAP analysis revealed that graph-theoretic features dominated the top-ranked predictors. These results demonstrate that graph-based topological features capture discriminative information in the ERG waveform that is inaccessible to conventional signal analysis methods, advancing the development of objective biomarkers for neurodevelopmental disorder screening. Full article

This paper presents a deterministic embedded control architecture for an industrial electroplating line. The validated system includes two autonomous trolleys, 18 station-aligned process positions, shared-track motion, and redundant grouped baths. The proposed controller addresses the limitations of rigid sequential automation by combining asynchronous finite-state trolley execution, runtime allocation of equivalent technological stations, dwell-time-preserving retrieval, distributed thermal supervision, and layered fail-safe protection within a single ATmega2560-based implementation. The core contribution is the integration of virtual process groups and temporal FIFO logic into a compact plant-side embedded controller. This enables adaptive bath selection and process-completion-based retrieval without reliance on a real-time operating system or a computationally heavy supervisory runtime. The architecture also incorporates predictive pre-start validation, runtime software arbitration, hardware-wired interlocks, binary-coded trolley positioning, and a distributed 1-Wire thermal measurement network. Validation was performed in a controller-centered hardware-in-the-loop representation of an 18-station zinc electroplating line. Over a 100-batch horizon, the proposed architecture reduced makespan from 1642 min to 1244 min, corresponding to a 24.2% throughput improvement. Average trolley idle time decreased from 18.4 min/batch to 4.1 min/batch. Grouped-bath utilization increased from 64% to 91%, while tracked bottleneck incidents decreased from 18 to 2. These results show that adaptive, resource-aware, and safety-layered electroplating control can be realized effectively on a compact embedded platform in an industry-representative HIL setting, while preserving dwell-time integrity and controller-level safety invariants. Full article

This study addresses the critical need for sustainable energy by optimizing diaphragm-type piezoelectric elements for efficient waste vibration energy harvesting. Traditional experimental optimization of complex, non-linear design parameters including applied load, tapper diameter, and support structures is often resource-intensive and time-consuming. To overcome these limitations, we developed a novel hybrid machine learning framework that seamlessly integrates an Artificial Neural Network (ANN) with a Modified Grey Wolf Optimization (mGWO) algorithm. The ANN was rigorously trained on experimental data using Bayesian Regularization, establishing itself as a robust and high-fidelity surrogate model capable of accurately predicting voltage output based on diverse input parameters, evidenced by an R-value close to 1. This predictive model subsequently served as the fitness function for the mGWO algorithm, which incorporated a non-linear control parameter to efficiently explore the multi-dimensional design space and effectively balance exploration with exploitation. The framework successfully identified the optimal configuration for maximizing voltage output, predicting a theoretical maximum of approximately 70.67 V. This optimal setup notably involved a high applied load of 100 N, the 6CA multi-pointed tapper configuration, and the three-support boundary condition, which is consistent with the experimentally validated results. The computational findings demonstrated excellent agreement with empirical results while providing significantly higher resolution for design insights. This validated, predictive tool offers a substantial advancement for the future scaling and design optimization of piezoelectric energy harvesters, minimizing the need for extensive physical prototyping and ensuring efficient stress transfer without mechanical failure. Full article

Personalized itinerary planning for large-scale passengers under resource constraints is a critical challenge in enhancing the operational efficiency and service quality of cruise tourism. Traditional clustering methods, which primarily rely on geometric similarity, often fail to address the intricate coupling between passenger preferences and finite venue capacities, lacking predictive capability for the ultimate planning quality. To overcome these limitations, this study proposes a novel bio-inspired data-driven hybrid optimization framework for the cruise itinerary planning task unit partition. The framework innovatively integrates a Genetic Balanced Clustering Algorithm (GBCA) for multi-objective passenger grouping, Kernel Principal Component Analysis (KPCA) for feature extraction from preference data, an improved Adaptive Spiral Flying Sparrow Search Algorithm (ASFSSA) for hyperparameter optimization, and a Kernel Extreme Learning Machine (KELM) for data-driven prediction of itinerary planning quality. This synergy enables the framework to dynamically allocate venue capacities based on group preferences and optimize partitioning towards maximizing overall benefits, ensuring load balance and fairness. Extensive experiments on simulated cruise scenarios demonstrate that the proposed framework significantly outperforms conventional methods, improving segmentation quality by at least 40% while exhibiting superior convergence speed and stability. This work provides a scalable, intelligent solution for complex resource-constrained scheduling problems, showcasing the effective application of bio-inspired data-driven methodologies in engineering optimization. Full article

With the rapid advancement of information technology, the energy consumption of data centers has become a critical issue. Accurate cooling load prediction is essential for optimizing cooling system operations and improving energy efficiency. However, conventional models often struggle to capture the complex nonlinearities and multi-variable coupling effects inherent in data centers. To address the limitations of existing models in terms of training efficiency and generalization performance, this study proposes a cooling load prediction model that integrates the light gradient boosting machine (LightGBM) algorithm with Bayesian optimization. The model was validated using data generated from an EnergyPlus simulation of a representative medium-scale data center. Comparative analysis demonstrates that the proposed model surpasses naive benchmarks (T-1, T-24, and T-168) and other machine learning models (SVR, XGBoost, and LSTM), achieving superior performance with a Root Mean Squared Error (RMSE) of 4.3234 kW, R² of 0.9999, and Mean Absolute Percentage Error (MAPE) of 0.07%. A noise robustness analysis further reveals that the model maintains excellent performance under realistic uncertainties, achieving an R² above 0.99 and an RPD exceeding 12 even at high noise levels (SNR = 20 dB). The total runtime and Relative Prediction Deviation (RPD) were 33.45 s and 86.2685, respectively, indicating an excellent balance between computational efficiency and robust predictive reliability. The key contribution of this research is the effective integration of LightGBM and Bayesian optimization to provide a highly accurate and efficient tool for data center cooling load prediction. This approach offers a scientific foundation for the intelligent control of cooling systems and energy efficiency optimization in data centers, with direct practical implications for building energy management. Full article

This study develops a unified multiscale–machine learning framework to interpret and predict thermo-mechanical wear regime transitions in MWCNT- and nanoclay-reinforced bio-based epoxy composites. A physics-informed master wear formulation integrating real contact mechanics, geometry-dependent shear transfer, interfacial adhesion energetics, and fracture-controlled matrix detachment was combined with interpretable machine learning analytics on a unified tribological dataset. In the CNT system, increasing loading from 0.1 to 0.4 wt.% enhanced interfacial adhesion energy density from 0.00813 to 0.01906 J/m², resulting in a monotonic reduction in the wear rate from 0.00918 to 0.00613 mm³/N·m (~33% reduction). In contrast, nanoclay exhibited an optimum behavior, with a minimum wear at 0.25 wt.% (0.000093 mm³/N·m; 7.9% reduction vs. neat clay baseline), followed by deterioration at a higher loading due to dispersion loss. The unified probabilistic regime classification of low-wear conditions (k < 0.007 mm³/N·m) achieved an ROC − AUC = 0.9256 and balanced accuracy = 94.3%, with thermo-mechanical severity identified as the dominant regime-switching driver. Reinforcement identity significantly modulated regime stability, confirming distinct shear transfer (Carbon Nano Tubes(CNT)) and confinement/tribofilm (clay) mechanisms within a common mathematical framework. By enabling the durability-oriented design of bio-based tribological systems and extending component service life through predictive stability mapping, this work contributes to resource-efficient materials engineering and reduced lifecycle waste, supporting Sustainable Development Goals SDG 9 (Industry, Innovation and Infrastructure), SDG 12 (Responsible Consumption and Production), and SDG 13 (Climate Action). Full article

Interest in isolated electrical systems powered by renewable energy has driven the development of alternatives to traditional Wind–Diesel Systems (WDS) due to their unwanted emissions and regulatory constraints. In this context, clean and efficient hybrid architectures are needed to comply with regulations and ensure stable operation under variations in user load and wind generation. This paper proposes an integrated isolated hybrid system consisting of a fuel cell replacing the Diesel Generator (DG). To fulfil the role of the synchronous generator in the diesel-group, the fuel cell operates under a Grid-Forming (GFM) control scheme, acting as a virtual synchronous machine that establishes the system’s voltage and frequency. The main aim of the hybrid system is for the wind turbine to supply most of the active power to the loads, thereby minimising hydrogen consumption. A key challenge in these systems is maintaining power balance, particularly preventing reverse flows in the fuel cell system, which has less margin than the diesel generator. In this paper, a Dump Load (DL) quickly dissipates excess power and prevents reverse power conditions. Overall, the proposed system eliminates the need for diesel generation, thereby eliminating emissions while maintaining operational stability. Simulation results demonstrate the correct functioning of the system in the presence of significant variations in load and wind power generation. Full article

The commercialization of V-based catalysts for the selective catalytic reduction of NOx by NH₃ (NH₃-SCR) is hindered by their narrow operating temperature window, insufficient low-temperature (LT) activity, and severe SO₂-to-SO₃ oxidation. To bridge this gap, we herein introduced Nb and hexagonal BN into a VW/TiO₂ system to simultaneously enhance its LT SCR activity, suppress undesired side reactions, and improve durability. Nb incorporation promoted V⁵⁺/V⁴⁺ redox cycling and enhanced lattice oxygen mobility, thus reducing the apparent activation energy and suppressing SO₂ oxidation at elevated temperatures. However, excessive Nb loading induced NH₃ oxidation and N₂O formation. This drawback was mitigated by introducing BN as a dispersion promoter, which helped secure high catalytic performance at a reduced Nb content. The VWNb/Ti-BN catalyst achieved superior NOx conversion and N₂ selectivity over a wide temperature range and benefited from notably suppressed NH₃ oxidation and SO₂-to-SO₃ oxidation. Kinetic analysis revealed that Nb primarily lowered the reaction energy barrier via redox property enhancement, whereas BN accelerated surface reaction turnover by stabilizing and dispersing active acidic sites, markedly increasing the turnover frequency without reducing the activation energy. In situ spectroscopic analysis confirmed the accelerated consumption of adsorbed NH₃ species and enhanced formation of reactive NOx intermediates, indicating SCR pathway enhancement. After aging in the presence of SO₂ and H₂O, the best-performing honeycomb-type monolithic catalyst retained and NOx conversion of >80%, demonstrating excellent long-term durability under practical conditions. A composition-aware machine learning model based on log-ratio-transformed variables quantitatively identified the synergistic balance among V, Nb, W, BN, and TiO₂ as the dominant factor governing LT SCR performance. Thus, this work provides valuable mechanistic insights and a strategy for designing wide-temperature-window SCR catalysts with improved activity, selectivity, and resistance to sulfur poisoning. Full article

As the number of electric vehicles (EV) increases, the demand for recommending the best charging location when using a large-scale charge network to charge is also increasing. A successful recommendation will utilize the user’s preference and the operational constraints of the charging network to make sure that it also takes into account the real-time operational requirements of the network. Most current papers focus on optimizing individual algorithmic components in isolation; consequently, many of these papers neglect to provide a holistic view of an integrated system. In addition, there are many operational requirements that current research does not consider, such as cold-start personalization for new users and enforcing real-time operational constraints like station availability, power capacity, maintenance windows, etc. This paper describes a deployable multi-stage recommendation system that creates a candidate list based on location and ranks preferences based on user, station and context features. The recommendation system also adds a configurable rule-based re-ranking layer to ensure that both hard constraints (i.e., charger availability and power-cap limits) and soft objectives (i.e., load balancing and operator priority) are enforced. A method for enabling mixed use between stable Bayesian and adaptive Bayesian methods was developed to provide users starting with cold-start performance that do not have adequate histories. Evaluation of this method using 100k+ real charging sessions showed that the fraction of sessions where the ground-truth station appears in the top-two recommendations (Hit@2) for the recommendation system was 0.82, representing a 37% increase in performance compared to proximity-based recommendation methods. The online deployed recommendation system has a 99th-percentile serving latency (P99) of less than 200 ms. The findings of this paper provide a framework for the implementation of operationally-relevant user-centric recommendation systems for EV services at scale. Full article