From Detection to Action: Using LLM Agents for Fault-Tolerant Control

2026-06-26

AI Summary

This work addresses the challenge of efficiently translating fault detection outcomes into fault-tolerant control actions that satisfy system constraints and are dynamically feasible. To this end, the authors propose a novel multi-agent framework grounded in large language models (LLMs), which uniquely integrates semantic knowledge graphs with collaborative multi-agent decision-making. The framework leverages a Digital Process Plant Twin (DPPT), CPSMod ontology-based Graph Retrieval-Augmented Generation (Graph RAG), finite-state-machine-guided recovery path planning, and a deterministic action validation mechanism to automatically generate safe recovery strategies from fault awareness. It provides unified support for both discrete and continuous process systems. Simulations on a hybrid batch processing unit and a continuous stirred-tank reactor demonstrate that lightweight LLMs can produce effective recovery actions within the latency bounds permitted by process dynamics, confirming the approach’s practicality and feasibility.

Abstract

We propose an agentic Large Language Model (LLM) framework for active Fault-Tolerant Control (FTC) that transforms fault detection outputs into constraint-aware recovery actions grounded in plant-specific knowledge. The approach couples (i) a multi-agent workflow that decomposes operator duties into monitoring, planning, action synthesis, simulation, validation, and reprompting; (ii) a Digital Process Plant Twin (DPPT) that exposes plant data, models, and a simulation service for pre-execution testing; and (iii) a Graph Retrieval-Augmented Generation (Graph RAG) layer built on the CPSMod ontology, which organizes plant knowledge (structure, function, hybrid dynamics, control context, and fault semantics) into a graph that supports relation-aware, multi-hop retrieval for the agents. Corrective actions are generated as minimal-risk state-machine recovery paths and corresponding discrete commands or continuous setpoint adaptations, then validated deterministically against interlocks, envelopes, and dynamic feasibility before any actuation. If no acceptable plan is found within a bounded time window, control is handed to a safety fallback. The framework is evaluated in simulation on two representative benchmarks: a discrete batch Mixing Module and a Continuous Stirred-Tank Reactor (CSTR) under closed-loop PID regulation. Results with lightweight LLMs (GPT-4o-mini and GPT-4.1-mini) show that semantically grounded agents can derive valid recovery decisions within latency budgets compatible with the respective process dynamics, demonstrating a practical pathway from detection to validated corrective action across both discrete and continuous FTC tasks.

Problem

• Fault-Tolerant Control
• LLM Agents
• Recovery Actions
• Process Plant
• Fault Detection

Innovation

• Fault-Tolerant Control
• LLM Agents
• Graph RAG
• Digital Twin
• Recovery Action Synthesis