Similar papers
Aug 2026
How effective are physics-informed neural networks compared to machine learning in predicting groundwater flow in complex heterogeneous aquifers?
This study explores the applicability of physics-informed neural networks to synthetic and real-world groundwater case studies, encompassing heterogeneous and homogeneous aquifers with varying boundary conditions and transient states, and reveals that PINNs provide a compelling alternative to classic ML methods, partic...
M. Bajpai, Shreyansh Mishra, S. Gaur et al.
· Neural computing & applicati... · 0 citations
Conference
Sep 2026
A hybrid model for agricultural pollution load forecasting by integrating meteorological data and soil properties: an XGBoost-LSTM approach
A hybrid forecasting model that fuses eXtreme Gradient Boosting for spatial feature importance evaluation with Long Short-Term Memory (LSTM) networks for sequential load prediction is proposed that provides a robust tool for proactive nutrient runoff management in data-sparse agricultural contexts.
Sun-Nan Meng, Sheng-Jun Jin, Hao Wang et al.
· International Conference on... · 0 citations
Review
Open access
Aug 2026
Next-Generation Approaches for Predicting Soil Heavy Metal Contamination: Insights from Geospatial and AI-Based Methods
The interrelation between geospatial and AI approaches represents a major breakthrough in environmental monitoring, offering a more in-depth and efficient tool for regulating heavy-metal pollution in soil.
Mingbao Zhu
· Journal of Environmental &am... · 0 citations
Open access
Sep 2026
Physically aligned forest fire risk prediction: A deep learning framework coupling fuel and climate multivariate factors
This paper proposes a physically aligned prediction framework named FWI-MSNet, using 18 years of synchronized observation data from the Huitong Ecological Station in China to construct a multi-scale feature system, selecting 21 physically relevant key features, including core indicators of the Forest Fire Weather Index...
Bo-Jie Chen, Anping Zeng, Yu Xie et al.
· PLoS ONE · 0 citations
Conference
Open access
2026
Towards physics-consistent machine learning models: A geomechanics-based artificial neural network for high-cyclic soil response
It is proposed to design a Geomechanics-based Artificial Neural Network (GANN) that bypasses the need for calibration parameters and instead uses common soil descriptors, ensuring that the predicted strain evolution remains consistent with soil mechanics principles.
R. Polo-Mendoza, M. Tafili, Jose Duque et al.
· E3S Web of Conferences · 0 citations
Open access
Sep 2026
Physics-Informed Neural Networks for One-Dimensional Groundwater Contaminant Transport: A Synthetic Numerical Study of Prediction and Parameter Inversion
The results show that PINNs accurately predict concentration values and reduce initialization-induced uncertainty when the physical constraints—including the ADE residual and the prescribed initial and boundary conditions—cover the target prediction period, outperforming purely data-driven neural networks in both accur...
Jiang-Wei Zhang, Wei Chen
· Water · 0 citations