Summary
This paper introduces an automated framework for constructing Dynamic Master Logic (DML) models as knowledge graphs (KG-DML) directly from system descriptions. DML offers a hierarchical representation linking functional objectives to underlying structural components, but conventional construction relies on expert interpretation of technical documentation, limiting scalability. The proposed approach combines retrieval-augmented generation with large language models to scale KG-DML construction to larger, more complex systems than prior work. Models are built layer by layer within the DML hierarchy using goal retrieval, preserving both functional dependencies and explicit logical relationships. The resulting KG-DML supports diagnostic reasoning, safety assessment, upward fault propagation, and downward dependency tracing. A multi-level validation strategy measures layer-specific precision and recall, logical gate consistency, and overall structural integrity. A case study on the Low Pressure Coolant Injection system of a decommissioned Boiling Water Reactor demonstrates consistent reconstruction across repeated runs. Results indicate that automated KG-DML can convert technical documentation into executable functional models suitable for diagnostics and reliability analysis.
Overview
Research area: Machine Learning (ML)
Authors: Saman Marandi, Yu-Shu Hu, Mohammad Modarres
Published: 2026-08-13
arXiv: 2508.03416
Summary
Dynamic Master Logic (DML) provides a hierarchical framework for representing system behavior by linking functional objectives to underlying structural components. However, building DML models typically depends on expert interpretation of technical documentation, which limits scalability for complex systems.
This study proposes a framework that automatically constructs DML models from system descriptions and represents them as knowledge graphs (KG-DML), using retrieval-augmented generation (RAG) and large language models (LLMs) as enabling tools. Building on prior work with smaller systems, the framework extends automated KG-DML construction and evaluation to larger and more complex systems.
Model construction proceeds layer by layer within the DML hierarchy, using goal retrieval to simultaneously preserve functional dependencies and explicit logical relationships. The resulting KG-DML supports diagnostic reasoning, safety assessment, upward fault propagation, and downward dependency tracing.
A multi-level validation method evaluates:
- Layer-specific precision and recall
- Logical gate consistency
- Overall structural integrity
Applied to the Low Pressure Coolant Injection (LPCI) system of a decommissioned Boiling Water Reactor, the case study demonstrates consistent reconstruction across repeated runs. The results suggest that automated KG-DML construction can transform technical documentation into executable functional models for diagnostic and reliability analysis.
Original Abstract
Dynamic Master Logic (DML) provides a hierarchical framework for representing system behavior...
---
*Automatically collected on 2026-08-14*
#paper #arXiv #ML
This page is an English static mirror generated for search and AI citation.
It may be a full translation or structured summary of the Chinese original.
Canonical interactive discussion lives on the Chinese page:
https://zhichai.net/topic/178633456