AgenticTwin: An Agentic LLM Framework Integrated with Digital Twin for Anomaly Detection

2026-08-12

AI Summary

This work addresses the challenge of interpreting anomaly detection results in digital twin systems, where the complexity and volume of sensor data often hinder operator comprehension. To this end, it proposes the first explainable anomaly analysis framework that integrates embodied agent architecture with digital twins, enabling natural language querying and generating interpretable diagnostics through structured agent collaboration, knowledge-anchored reasoning, and context-aware retrieval. The study innovatively introduces synthetic anomaly injection and query generation mechanisms to establish a rigorous benchmark for evaluation and demonstrates the feasibility of deploying lightweight, open-source large language models in real-world cyber-physical systems. Experimental results on a meteorological sensor dataset show that the proposed approach significantly outperforms baseline methods in diagnostic quality, retrieval accuracy, and the effectiveness of mitigation recommendations.

Abstract

Digital twins are increasingly used to monitor and simulate the behavior of cyber-physical systems. Even with skilled operators, interpreting anomalies detected within digital twin pipelines is challenging, as the sheer complexity and volume of raw sensor data make thorough analysis difficult. Recent advances in large language models (LLMs) offer promising capabilities for reasoning and explanation, yet their integration into digital twin-driven anomaly analysis remains underexplored. In this work, we propose AgenticTwin, an agentic framework that integrates LLM-driven reasoning with a digital twin-based anomaly detection pipeline. The framework grounds LLM-generated explanations in outputs from a digital twin-driven anomaly classifier and enables human operators to ask relevant natural-language questions about the system. Beyond the framework itself, we introduce a benchmark-oriented evaluation pipeline constructed over synthetic anomalies injected into a real-world weather sensor dataset, enabling controlled generation of operator queries over anomaly events. We further evaluate the feasibility of deploying lightweight, open-source LLMs for practical cyber-physical environments. Experimental results demonstrate that structured agent collaboration and knowledge-grounded reasoning improve diagnosis quality, contextual retrieval, and mitigation quality across diverse possible anomaly scenarios.

Problem

• digital twin
• anomaly detection
• cyber-physical systems
• sensor data interpretation
• large language models

Innovation

• Agentic LLM
• Digital Twin
• Anomaly Detection
• Knowledge-Grounded Reasoning
• Cyber-Physical Systems