The field of physics-informed neural networks (PINNs) and control systems is rapidly advancing, with a focus on developing innovative methods for solving complex problems in various domains. Recent research has explored the application of PINNs to areas such as turbulence control, pandemic modeling, and material optimization. A key trend in this field is the integration of physical laws and constraints into neural network architectures, enabling more accurate and efficient solutions to forward and inverse problems. Another notable direction is the development of novel control strategies, including model predictive control and reinforcement learning, which are being applied to complex systems such as roll-to-roll manufacturing and turbulent flows. Noteworthy papers in this area include the work on Physics-Informed Neural-operator Predictive Control for Drag Reduction in Turbulent Flows, which achieved a drag reduction of 39.0% using a PINO-PC approach, and the paper on Learning Pareto-Optimal Pandemic Intervention Policies with MORL, which demonstrated the effectiveness of a multi-objective reinforcement learning framework for modeling and evaluating disease-spread prevention strategies.
Advances in Physics-Informed Neural Networks and Control Systems
Sources
Dynamic Modeling and Control System Analysis for Continuous-Disc Filters in Pulp Mill Operations
Deducing Closed-Form Expressions for Bright-Solitons in Strongly Magnetized Plasmas with Physics Informed Symbolic Regression (PISR)
Quantifying constraint hierarchies in Bayesian PINNs via per-constraint Hessian decomposition
Learning Pareto-Optimal Pandemic Intervention Policies with MORL
Physics-informed Neural-operator Predictive Control for Drag Reduction in Turbulent Flows
Training Variation of Physically-Informed Deep Learning Models
A Complement to Neural Networks for Anisotropic Inelasticity at Finite Strains
Towards Fast Option Pricing PDE Solvers Powered by PIELM
Data-Driven Adaptive PID Control Based on Physics-Informed Neural Networks
Deep vs. Shallow: Benchmarking Physics-Informed Neural Architectures on the Biharmonic Equation
Data-driven linear solver selection and performance tuning for multiphysics simulations in porous media
Model Predictive Path Integral Control for Roll-to-Roll Manufacturing
StruSR: Structure-Aware Symbolic Regression with Physics-Informed Taylor Guidance
AutoBalance: An Automatic Balancing Framework for Training Physics-Informed Neural Networks
Modeling COVID-19 Dynamics in German States Using Physics-Informed Neural Networks
TOMATOES: Topology and Material Optimization for Latent Heat Thermal Energy Storage Devices