A backtracking search-based extreme gradient boosting algorithm for soil moisture prediction using meteorological variables

Earth Science Informatics, Journal Year: 2025, Volume and Issue: 18(2)

Published: Jan. 16, 2025

Language: Английский

---

Adjusting Soil Temperatures with a Physics-Informed Deep Learning Model for a High-Resolution Numerical Weather Prediction System

Atmosphere, Journal Year: 2025, Volume and Issue: 16(2), P. 207 - 207

Published: Feb. 12, 2025

Soil temperature (ST) plays an important role in the surface heat energy balance, and accurate description of soil temperatures is critical for numerical weather prediction; however, it difficult to consistently measure temperatures. We developed a U-Net-based deep learning model derive (designated as ST-U-Net) primarily based on 2 m air (T2) forecasts. The model, domain which covers Mt. Lushan region, was trained tested by utilizing high-resolution forecast archive operational research forecasting four-dimensional data assimilation (WRF-FDDA) system. results showed that ST-U-Net can accurately estimate T2 inputs, achieving mean absolute error (MAE) less than 0.8 K testing set 5055 samples. performance varied diurnally, with smaller errors at night slightly larger daytime. Incorporating additional inputs such land uses, terrain height, radiation flux, coded time further reduced MAE ST 26.7%. By developing boundary-layer physics-guided training strategy, 8.8%.

Language: Английский

---

Machine learning modeling and multi objective optimization of artificial detrusor

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: April 22, 2025

To address the problem of obtaining optimal design parameters for existing artificial detrusors using single-objective optimization methods, this research proposed a machine learning-based detrusor modeling and multi-objective approach, which includes thorough process from to optimization. Extreme learning was used model in suggested order increase accuracy, multi-strategy modified crayfish algorithm tuning extreme machine's put forth research. The grey wolf utilized optimize based on model. In validate an driven by shape memory spring finally built as experimental platform. results show that improved paper can effectively avoid defects original algorithm, its performance convergence ability are better than comparison algorithm. With root mean square error 1.51E-02, coefficient determination 9.81E-01, absolute 1.32E-02, percentage 1.66E-01, established predicts spring-driven detrusor's emptying rate. It also temperature increment with 8.47E-01, 5.81E-01, 7.23E-02. These predictions superior prediction model, indicating good predictive stability. Additionally, demonstrates outstanding uncertainty reliability analysis, thereby further confirming comprehensive performance. optimized computed values rate increment, well measurement values, have errors 7.8% 11.8%, respectively, satisfy engineering specifications. our method exhibits significant enhancements over designs. Specifically, achieves approximately 20% 62% reduction successfully balancing urinary efficiency mitigated risks thermal tissue injury.

Language: Английский

---

From blender to farm: Transforming controlled environment agriculture with synthetic data and SwinUNet for precision crop monitoring

PLoS ONE, Journal Year: 2025, Volume and Issue: 20(4), P. e0322189 - e0322189

Published: April 24, 2025

The aim of this study was to train a Vision Transformer (ViT) model for semantic segmentation differentiate between ripe and unripe strawberries using synthetic data avoid challenges with conventional collection methods. solution used Blender generate strawberry images along their corresponding masks precise segmentation. Subsequently, the were evaluate SwinUNet as method, Deep Domain Confusion utilized domain adaptation. trained then tested on real from Strawberry Digital Images dataset. performance achieved Dice Similarity Coefficient 94.8% 94% strawberries, highlighting its effectiveness applications such fruit ripeness detection. Additionally, results show that increasing volume diversity training can significantly enhance accuracy each class. This approach demonstrates how datasets be employed cost-effective efficient overcoming scarcity in agricultural applications.

Language: Английский

---

Earth-sheltered buildings: A review of modeling, energy conservation, daylighting, and noise aspects

Journal of Cleaner Production, Journal Year: 2024, Volume and Issue: 472, P. 143482 - 143482

Published: Aug. 25, 2024

Buildings constitute approximately 40% of global energy consumption and contribute to around one-third total greenhouse gas emissions. Buried, earth-sheltered, slope buildings leverage the ground's thermal mass conserve heating cooling energy. This review examines modeling performance in earth-sheltered buildings, focusing on temperature profile ground summarizing published mathematical formulations solutions for distribution. It encompasses considerations both earth surface measurements balance at earth's surface. Representative boundary conditions their parameters, including conductivity, heat transfer, radiation intensity, temperature, humidity metrics, are summarized. Efforts model behavior as well soil various depths, using statistical machine learning methods, artificial neural networks, also documented. The further explores conservation location studies, methodology, construction description achieved efficiency, confirming significant role a consequent reduction requirements. Subsequently, daylighting is examined, windows, impact circadian rhythm, architectural such skylights courtyards. A noise aspects considers shielding from airborne ground-borne exterior sources, transmission losses, structure-borne vibrations. Indoor acoustic environments, with measures quality like reverberation time speech transmission, discussed. Sound absorption examined indoor caves, rock-cut structures, underground buildings. Finally, health safety issues refer ventilation, lighting, structural safety. concludes synthesis findings.

Language: Английский

---

Efficient Computational Investigation on Accurate Daily Soil Temperature Prediction Using Boosting Ensemble Methods Explanation Based on SHAP Importance Analysis

Results in Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 103220 - 103220

Published: Oct. 1, 2024

Language: Английский

---

Predicting the drag coefficient of coastal trees using Support Vector Machines and boosting ensemble models

Discover Water, Journal Year: 2024, Volume and Issue: 4(1)

Published: Nov. 13, 2024

The effect of green belts on wave absorption is a critical aspect coastal protection strategies. effectiveness in influenced by factors such as the type vegetation used, density and width belt, topography coastline. current study aims to explore performance various intelligent tools, including SVM (Support Vector Machine), ABR (Ada Boost Regression), ETR (Extra Trees GBR (Gradient Boosting RF (Random Forest), forecast drag coefficients trees (CD). In this direction, four dimensionless parameters relative height (H/d), (D), shoreline slope (S), propagation velocity (u/ $$\sqrt{\xi E/\rho }$$ ) were assumed input parameters, CD was considered target. To evaluate developed soft computing models, statistical indicators graphical plots Violin, Tylor, Scatter applied. results revealed that method outperforms existing machine learning techniques with R2 = 0.996, RMSE 0.003, MAE 0.002, SI 0.014. addition, Tylor diagram indicates distance index obtained using model exhibited high alignment actual data, especially comparison alternative tools.

Language: Английский

---

Retrieval of global surface soil and vegetation temperatures based on multisource data fusion

Remote Sensing of Environment, Journal Year: 2024, Volume and Issue: 318, P. 114564 - 114564

Published: Dec. 13, 2024

Language: Английский

---

Soil temperature estimation at different depths using machine learning paradigms based on meteorological data

Environmental Monitoring and Assessment, Journal Year: 2024, Volume and Issue: 197(1)

Published: Dec. 26, 2024

Language: Английский

---