Pouyan Nasiri | Searcharxiv

arXiv2025

Data-Driven Model for Elastomers under Simultaneous Thermal and Radiation Exposure

We present a physics-informed neural network framework for predicting the mechanical performance of elastomers exposed to concurrent thermal and gamma-radiation exposure, such as elastomers in nuclear cables or space electronics. Our demonstrated approach integrates the dual-network hypothesis with the microsphere concept to represent soft and brittle sub-networks, while embedding physical laws directly into the machine learning process. Hard constraints, e.g., incompressibility, bounded network fractions are enforced through network architecture, and soft constraints e.g., monotonicity, polyconvexity, and fading effects are imposed through the loss function. This integration reduces the effective search space, guiding the optimization toward physically admissible solutions and enhancing robustness under sparse data. Validation against published datasets on silicone rubber, ethylene propylene diene monomer, and silica-reinforced silicone foam shows accurate predictions of stress-strain behavior and elongation-at-break at exposure times not used for training. Results confirm that physics-informed constraints improve extrapolation, capture synergistic thermal-radiation effects, and provide a reliable tool for lifetime assessment of nuclear cable insulation and other radiation-exposed elastomers.

physics.gen-ph↗

arXiv2022

Reduced-PINN: An Integration-Based Physics-Informed Neural Networks for Stiff ODEs

Physics-informed neural networks (PINNs) have recently received much attention due to their capabilities in solving both forward and inverse problems. For training a deep neural network associated with a PINN, one typically constructs a total loss function using a weighted sum of different loss terms and then tries to minimize that. This approach often becomes problematic for solving stiff equations since it cannot consider adaptive increments. Many studies reported the poor performance of the PINN and its challenges in simulating stiff chemical active issues with administering conditions of stiff ordinary differential conditions (ODEs). Studies show that stiffness is the primary cause of the failure of the PINN in simulating stiff kinetic systems. Here, we address this issue by proposing a reduced weak-form of the loss function, which led to a new PINN architecture, further named as Reduced-PINN, that utilizes a reduced-order integration method to enable the PINN to solve stiff chemical kinetics. The proposed Reduced-PINN can be applied to various reaction-diffusion systems involving stiff dynamics. To this end, we transform initial value problems (IVPs) to their equivalent integral forms and solve the resulting integral equations using physics-informed neural networks. In our derived integral-based optimization process, there is only one term without explicitly incorporating loss terms associated with ordinary differential equation (ODE) and initial conditions (ICs). To illustrate the capabilities of Reduced-PINN, we used it to simulate multiple stiff/mild second-order ODEs. We show that Reduced-PINN captures the solution accurately for a stiff scalar ODE. We also validated the Reduced-PINN against a stiff system of linear ODEs.

cs.LG↗