Real-time Health Monitoring of Heat Exchangers using Hypernetworks and PINNs
This work addresses predictive maintenance for heat exchangers in digital twin environments, offering a domain-specific incremental improvement over existing PINNs.
The paper tackles real-time health monitoring of heat exchangers in thermal power plants by developing a hypernetwork-based Physics-informed Neural Network (PINN) model that learns thermal behavior under dynamic conditions without retraining, achieving orders of magnitude reduction in inference time while maintaining accuracy comparable to physics-based simulations.
We demonstrate a Physics-informed Neural Network (PINN) based model for real-time health monitoring of a heat exchanger, that plays a critical role in improving energy efficiency of thermal power plants. A hypernetwork based approach is used to enable the domain-decomposed PINN learn the thermal behavior of the heat exchanger in response to dynamic boundary conditions, eliminating the need to re-train. As a result, we achieve orders of magnitude reduction in inference time in comparison to existing PINNs, while maintaining the accuracy on par with the physics-based simulations. This makes the approach very attractive for predictive maintenance of the heat exchanger in digital twin environments.