Physics-Informed Neural Network Modeling of Biodegradable Contaminant Transport through GCL/SL Composite Liners
Researchers have developed a novel physics-informed neural network (PINN) framework to model contaminant transport through composite liners. This framework, particularly the hard-constrained PINN (H-PINN) variant, significantly improves accuracy and stability in predicting contaminant concentrations compared to standard PINNs. The H-PINN also demonstrates effectiveness in inverse modeling, successfully identifying degradation half-lives from limited observational data. AI
IMPACT This research introduces a more accurate method for modeling contaminant transport, potentially improving environmental engineering and risk assessment.