Fast Bayesian equipment condition monitoring via simulation based inference: applications to heat exchanger health

arXiv cs.LG / 4/23/2026

📰 NewsDeveloper Stack & InfrastructureSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper proposes an AI-driven Simulation-Based Inference (SBI) framework to infer latent degradation parameters of industrial heat exchangers from uncertain sensor observations.
Instead of using computationally expensive Bayesian inference via MCMC, it trains amortized neural density estimators to learn a direct, likelihood-free mapping from thermal-fluid measurements to the full posterior over degradation parameters.
Experiments on synthetic fouling and leakage scenarios—including rare, low-probability sparse-event failures—show SBI matches MCMC in diagnostic accuracy and provides reliable uncertainty quantification.
The method substantially speeds up inference, reporting an approximately 82× reduction in time versus traditional sampling, enabling near-instantaneous diagnosis suitable for real-time control and digital twin applications.

Abstract

Accurate condition monitoring of industrial equipment requires inferring latent degradation parameters from indirect sensor measurements under uncertainty. While traditional Bayesian methods like Markov Chain Monte Carlo (MCMC) provide rigorous uncertainty quantification, their heavy computational bottlenecks render them impractical for real-time process control. To overcome this limitation, we propose an AI-driven framework utilizing Simulation-Based Inference (SBI) powered by amortized neural posterior estimation to diagnose complex failure modes in heat exchangers. By training neural density estimators on a simulated dataset, our approach learns a direct, likelihood-free mapping from thermal-fluid observations to the full posterior distribution of degradation parameters. We benchmark this framework against an MCMC baseline across various synthetic fouling and leakage scenarios, including challenging low-probability, sparse-event failures. The results show that SBI achieves comparable diagnostic accuracy and reliable uncertainty quantification, while accelerating inference time by a factor of82

\times

compared to traditional sampling. The amortized nature of the neural network enables near-instantaneous inference, establishing SBI as a highly scalable, real-time alternative for probabilistic fault diagnosis and digital twin realization in complex engineering systems.