Applied Modelica for Autonomous Cyber Defense: Engineer multi-agent workflows to protect industrial control systems

By Bernand Bernie

Defend Cyber-Physical Systems with Agentic Artificial Intelligence

Industrial control systems are under constant threat. Standard network security is no longer sufficient to protect the physical machinery that runs modern infrastructure. You must construct defensive architecture that natively understands both the IT network and the physical plant.

Applied Modelica for Autonomous Cyber Defense provides a complete, hands-on blueprint for engineering multi-agent artificial intelligence workflows that monitor, diagnose, and defend industrial systems. By combining the physical simulation power of OpenModelica with the orchestration capabilities of LangGraph and Claude, you will build a sophisticated autonomous defense network.

This project-first manual prioritizes functional code implementation over abstract theory. You will build a complete physical simulation of an industrial pump station, connect it to a Java microservices control layer, and deploy Wazuh for threat detection. From there, you will code diagnostic and response AI agents that collaborate to detect IT intrusions and physical sabotage in real time.

What You Will Build:

A high-fidelity physical plant simulation using OpenModelicaA secure Java enterprise microservices backendCustom LangGraph tool APIs for reading sensor data and Wazuh logsA vector memory database using Pgvector for historical threat retrievalAutonomous AI agents that isolate compromised services and trigger safe plant shutdowns

Protect the digital and physical boundaries of your infrastructure. Engineer your autonomous defense system today.

• Format: Paperback
• Language: English
• ISBN: B0H44PD1N5
• ISBN13: 9798199991643
• Release Date: June 2026
• Publisher: Independently Published
• Length: 242 Pages
• Weight: 0.94 lbs.
• Dimensions: 0.5" x 7.0" x 10.0"