Physics-informed neural network predicts slope reliability under variable conditions
Label-free physics-informed strength reduction and a neural operator for the reliability of spatially variable slopes
Computational Engineering, Finance, and Science
Summary
Slopes like hills can fail or slide unpredictably, especially when their material properties vary across space. The authors developed a special neural network that uses physics rules to estimate how stable these slopes are without needing lots of labeled data. Their method runs many simulations quickly on a graphics card and matches results from standard engineering software closely. They also created another neural model that can predict slope failure risks and behavior very fast across millions of scenarios.
What this means in practice
- •For geotechnical engineers: Estimate slope failure risks and failure modes under varying soil conditions quickly without running multiple finite-element analyses.
- •For risk assessment teams: Perform large-scale probabilistic simulations of slope stability scenarios using a trained neural operator, accelerating reliability evaluations.
Authors
Apisit Robjanghvad, Sompote Youwai
Abstract
This paper proposes a label-free physics-informed neural network (PINN) for non-associated Mohr-Coulomb plasticity with incremental strength reduction for the reliability of spatially variable $c$-$φ$ slopes. The PINN is trained with a virtual-work loss on the finite-element mesh and solves many random fields at once on a graphics processing unit. With a solver budget fixed on the homogeneous slopes, it reproduces the Abaqus factor of safety within 2 to 3% root-mean-square on 52 random fields and locates the failure mechanism in the same weak zones, with the plastic strain in the band under-resolved. A neural operator trained only on the PINN solutions predicts the factor of safety, displacement path and failure mode of a million fields in seconds. No finite-element solution enters the pipeline: Monte Carlo fields, importance samples and operator training data come from the PINN, and Abaqus serves only to verify the result. Monte Carlo simulation, the first-order reliability method, importance sampling and the operator give consistent failure probabilities down to $10^{-3}$ for the slope studied, but a 3% scatter in the factor of safety inflates a probability of $10^{-3}$ by up to a factor of two, and the absolute values refer to one mesh. A heterogeneous Bishop analysis gives failure probabilities 2.6 and 8.3 times lower although its critical circles follow the PINN mechanism, because a small difference in the factor of safety is amplified in the tail. The operator also yields the failure mechanism and displacement response conditional on failure.