A method and system for secure integration of digital twin ioc system agents with devices

By decoupling the AI ​​large-scale intelligent agent from the device execution authority, introducing risk classification and human-machine collaborative approval mechanisms, and constructing OT fuses for physical rule verification, the security risks in the integration of AI large-scale intelligent agents and physical devices are solved, achieving a balance between safety and efficiency and closed-loop management of device status.

CN122431280APending Publication Date: 2026-07-21BEIJING DIGITAL HAIL TECH
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING DIGITAL HAIL TECH
Filing Date
2026-04-22
Publication Date
2026-07-21

Smart Images

  • Figure CN122431280A_ABST
    Figure CN122431280A_ABST
Patent Text Reader

Abstract

The application discloses a kind of digital twin IOC system intelligent agent and the security integration method and system of equipment, method includes: AI big model intelligent agent is decoupled with equipment execution right, AI big model intelligent agent is only responsible for natural language intention analysis and standardization operation request generation, does not hold physical equipment control authority;Risk grading engine is built, low-risk operation is directly executed, high-risk operation is forced into man-machine collaborative approval process;For high-risk operation, automatically generate approval card and push to authorized administrator through collaborative instant messaging platform, cool period anti-shake, safety threshold, mutual exclusion logic and operation frequency physical rule verification are carried out to device operation instruction;Device reverse state data is actively pushed to AI big model intelligent agent, realizes pre-checking and state linkage;Configuration security policy management center, build the depth security defense system from IT layer to OT layer, solve the misoperation and overreach execution risk when AI big model intelligent agent directly controls physical equipment.
Need to check novelty before this filing date? Find Prior Art