Hydrogen is widely recognized as a promising zero-carbon fuel for internal combustion engines due to its zero carbon emissions, high laminar flame speed, and wide flammability limits. In dual-fuel compression ignition engines, substituting part of the diesel fuel with hydrogen can significantly reduce carbon-based emissions while maintaining or improving ther-mal efficiency. This study investigates the combined effects of hydrogen substitution ratio (HSR) and equivalence ratio (Φ) on combustion characteristics, engine performance, and emissions in a diesel-hydrogen dual-fuel engine. Three HSR levels (50, 70, and 90) are examined across equivalence ratios ranging from 0.5 to 0.8. A three-dimensional computational fluid dynamics (CFD) model is developed using ANSYS Forte. The model employs an Eulerian-Lagrangian framework, incorpo-rating the unsteady Reynolds-averaged Navier–Stokes (URANS) equations for flow resolution and the G-equation for turbu-lent flame propagation. Spray breakup dynamics are modeled using the Kelvin–Helmholtz/Rayleigh–Taylor approach, while turbulence is captured with the RNG k-ε model. Results indicate that increasing both the HSR and equivalence ratio reduces CO₂ emissions. At HSR 70, unburned hydrocarbon (UHC) emissions decrease by up to 82% as Φ increases from 0.5 to 0.8. Furthermore, the indicated specific fuel consumption (ISFC) decreases, particularly at high HSR, reflecting improved fuel economy. These findings highlight the potential of diesel-hydrogen dual-fuel operation for achieving cleaner and more efficient engine performance.
Field hospitals are temporary healthcare facilities deployed to provide medical services in scenarios where conventional healthcare infrastructure is disrupted due to natural or human-induced disasters, migration, armed conflicts, or similar crises. Depending on the anticipated needs, these facilities may be categorized into specialized types, such as orthopedic and traumatology field hospitals, surgical field hospitals, or general medical aid field hospitals. Bed capacity is a critical determinant of the scale of a field hospital, which in turn influences the spatial requirements for setup, personnel allocation, and associated costs. Typically established for short-term use, ranging from six months to two years, the operational duration, bed capacity, and structural configuration of field hospitals collectively define their functional capabilities. This study proposes a two-stage stochastic mathematical programming model to address the problem of calculating the optimal bed capacity for field hospitals delivering post-disaster healthcare services, with a focus on earthquake scenarios. Compared with two-stage stochastic models, the mathematical model proposed in this study exhibits distinct differences from existing models in the disaster and hospital bed capacity modeling literature. The model integrates uncertainties inherent in disaster contexts, such as fluctuating patient influx and resource availability, to optimize decision-making for bed allocation. The proposed framework is applied to Avcılar, a district within Istanbul, Türkiye, identified as a high-risk zone for seismic activity due to its proximity to the North Anatolian Fault. By leveraging geospatial and demographic data, the model aims to enhance preparedness and operational efficiency of field hospitals in mitigating healthcare service disruptions following a major earthquake. The results underscore the utility of stochastic optimization in resource planning under uncertainty, offering actionable insights for policymakers and disaster response coordinators.
The increasing penetration of renewable energy introduces additional complexity into long-term transmission expansion planning because investment decisions must simultaneously account for network topology, nonlinear operating constraints, renewable-generation uncertainty, and contingency security. This study proposes a Graph Neural Network-assisted Proximal Policy Optimization (GNN-PPO) framework for multi-stage security-constrained transmission expansion planning (SC-TEP). A two-layer edge-aware GNN with a 16-dimensional embedding is employed to encode the transmission-network topology, while PPO sequentially selects reinforcement projects over the planning horizon. The framework incorporates AC power-flow constraints, three renewable-availability scenarios, candidate-specific investment costs, and N−1 security requirements. To reduce computational burden, the five most critical credible contingencies are screened during training, whereas all credible non-islanding N−1 contingencies are evaluated during post-training validation. A stage-aware security-repair procedure followed by backward project pruning is introduced to eliminate residual security violations and unnecessary reinforcements. The framework is applied to the Southern Sulawesi 150 kV transmission system over the 2025–2034 planning horizon. The raw GNN-PPO portfolio requires 64.17 million USD (MUSD), compared with 97.10 MUSD for the Multilayer Perceptron-based PPO (MLP-PPO) benchmark, but full-horizon validation identifies one residual contingency violation in 2025. After security repair and pruning, the final solution consists of eight transmission reinforcement projects with a total planning investment of 86.72 MUSD. In 2034, the final system achieves a minimum voltage of 1.0038 pu, maximum branch loading of 100.00%, transmission losses of 41.31 MW, and renewable curtailment of 495.42 MW, with zero failures among 62 credible non-islanding N−1 contingencies. The results demonstrate that topology-aware reinforcement learning can provide an effective candidate-search mechanism for multi-stage SC-TEP, while explicit post-training security verification remains essential for obtaining an engineering-feasible expansion plan.
From the perspective of global food security, the estimation of crop biomass and yield is a critical process. Traditionally, plant height is manually measured using field samples for biomass and yield estimation, and plant samples are cut, dried, and weighed. This labor-intensive process can be made more efficient through advanced technologies integrated into modern agricultural machinery. One such technology, light detection and ranging (LiDAR), provides nondestructive 3D point clouds with millimeter-level accuracy. However, processing this data is highly challenging due to the complex canopy architecture of cereal crops. Therefore, deep learning architectures can be used to effectively analyze 3D point cloud data. In this context, a deep learning-based regression approach was proposed to estimate above-ground biomass from 3D point cloud data. This study employed a point cloud transformer (PCT) architecture. Growth Stage-PCT (GS-PCT) models were developed by integrating growth stage information into the proposed Small, Base, and Deep variants of the PCT architecture. According to the experimental results, when both growth stages were evaluated together, the proposed GS-PCT-Small model achieved the best performance, with an R² of 0.95, an RMSE of 88.60 g m⁻², an MAE of 69.34 g m⁻², and a MAPE of 13.69%. During the vegetative stage, the PCT-Small model achieved an RMSE of 58.52 g m⁻², while during the flowering stage, the GS-PCT-Small model achieved an RMSE of 111.55 g m⁻². The findings indicate that the models achieved high predictive performance for above-ground biomass estimation from 3D point clouds.
In The increasing electrification of urban bus fleets necessitates maintenance strategies that reflect real-world operational variability rather than relying on fixed mileage intervals. This study proposes a Decision Tree (DT)-based dynamic maintenance interval model developed using approximately 58,343 high-resolution operational records collected from 20 drivers operating the same electric bus route under comparable boundary conditions. The dataset includes energy consumption, vehicle speed, vehicle mass, ambient temperature, road slope, acceleration, regenerative braking power, and traffic congestion indicators. Correlation analysis revealed strong associations between operational variables and failure indicators, with coefficients reaching up to 0.94. Among the evaluated regression techniques, the Decision Tree model demonstrated superior predictive performance (R² = 0.95, RMSE = 0.314, MSE = 0.1523, MAE = 0.343), outperforming Support Vector Machine (R² = 0.93) and Gaussian Process Regression (R² = 0.92). The optimal operating profile identified by the DT-based estimation indicated an energy consumption level of 501 kW, an average vehicle speed of 47 km/h, acceleration of 0.27 m/s², slope of 3.45%, and recuperation power of 612 kW under 23.5% traffic congestion. Driver-level comparisons showed that deviations from these optimal parameters resulted in measurable shifts in projected maintenance intervals, particularly for drivers exhibiting higher acceleration (≥0.35 m/s²) and elevated energy consumption (>540 kW). The findings demonstrate that maintenance thresholds in electric bus systems are strongly influenced by operational behavior and can be dynamically adjusted through data-driven modeling. The proposed framework offers a quantitative basis for predictive maintenance scheduling, enabling fleet operators to reduce premature servicing and mitigate failure risk while improving operational efficiency.
