FaultExplainer: Leveraging Large Language Models for Interpretable Fault Detection and Diagnosis

TL;DR

FaultExplainer is an interactive tool that uses large language models to improve interpretability and diagnosis of faults in chemical processes, integrating sensor data visualization and PCA-based detection.

Contribution

The paper introduces FaultExplainer, a novel LLM-powered system that enhances fault detection, diagnosis, and explanation with real-time data and interactive features in chemical process monitoring.

Findings

GPT-4o and o1-preview generate plausible explanations

System effectively identifies top contributing variables

Limitations include reliance on PCA features and hallucinations

Abstract

Machine learning algorithms are increasingly being applied to fault detection and diagnosis (FDD) in chemical processes. However, existing data-driven FDD platforms often lack interpretability for process operators and struggle to identify root causes of previously unseen faults. This paper presents FaultExplainer, an interactive tool designed to improve fault detection, diagnosis, and explanation in the Tennessee Eastman Process (TEP). FaultExplainer integrates real-time sensor data visualization, Principal Component Analysis (PCA)-based fault detection, and identification of top contributing variables within an interactive user interface powered by large language models (LLMs). We evaluate the LLMs' reasoning capabilities in two scenarios: one where historical root causes are provided, and one where they are not to mimic the challenge of previously unseen faults. Experimental results using GPT-4o and o1-preview models demonstrate the system's strengths in generating plausible and actionable explanations, while also highlighting its limitations, including reliance on PCA-selected features and occasional hallucinations.