Real-time Health Monitoring of Heat Exchangers using Hypernetworks and PINNs

TL;DR

This paper presents a hypernetwork-enhanced PINN model for real-time heat exchanger health monitoring, significantly reducing inference time while maintaining accuracy, enabling effective predictive maintenance in energy systems.

Contribution

It introduces a domain-decomposed hypernetwork approach for PINNs that eliminates re-training, achieving rapid inference for real-time thermal health monitoring.

Findings

Orders of magnitude faster inference compared to traditional PINNs

Maintains accuracy comparable to physics-based simulations

Suitable for digital twin predictive maintenance environments

Abstract

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.