Physics-Informed Neural Network Modeling of Biodegradable Contaminant Transport through GCL/SL Composite Liners
2026-06-03 • Machine Learning
Machine LearningComputation and Language
AI summaryⓘ
The authors created a special neural network to understand how contaminants move through a two-layer liner system used in landfills. They compared two versions: one that loosely follows certain rules (Std-PINN) and one that strictly follows those rules (H-PINN). The strict version gave more accurate and stable predictions, especially when contaminants move faster. They also showed it can help figure out how quickly the soil layer breaks down contaminants, even with limited and noisy data.
Physics-Informed Neural Network (PINN)Contaminant TransportGeosynthetic Clay Liner (GCL)Soil Liner (SL)Advection-Dispersion-BiodegradationInverse ModelingLeachate-HeadActivation FunctionMean Absolute Error (MAE)Mean Relative Error (MRE)
Authors
Dong Li, Yapeng Cao, Haiping Zhao, Shutong Han
Abstract
This study develops a two-domain physics-informed neural network framework for contaminant transport through a GCL/SL composite liner system, in which the thin GCL layer is treated using a steady-state advection-dispersion-biodegradation formulation and the underlying soil liner is modeled as a transient transport domain. Two formulations are evaluated against analytical and finite-element reference solutions under different leachate-head conditions: a standard PINN with soft constraint enforcement (Std-PINN) and a hard-constrained PINN (H-PINN), in which selected boundary and initial conditions are embedded directly into the trial solutions. The Std-PINN captures the overall breakthrough behavior but shows larger errors during the early transport stage, particularly under higher leachate heads where advective transport becomes more pronounced. The H-PINN reduces the optimization burden associated with penalty-based constraint enforcement and provides more accurate and stable concentration predictions, lowering the MAE from approximately 0.058-0.067 for the Std-PINN to about 0.011-0.023 for the H-PINN, while reducing the MRE from approximately 9.10%-19.16% to about 2.08%-3.14%. Parametric analyses confirm that the H-PINN with the tanh activation function and an optimized network structure provides the best predictive accuracy. The H-PINN is further extended to inverse modeling for identifying the SL degradation half-life from limited concentration observations, showing reliable convergence toward prescribed values and acceptable robustness under low-to-moderate observation noise.