An automated and adaptive test sequence generation system for EMI testing

By integrating domain knowledge and using online reinforcement learning to combine the spectrum data and test process data of the EMI test system, an adaptive test sequence is generated, which solves the problems of rigid test paths and insufficient completeness in existing technologies, and achieves efficient and complete EMI testing.

CN122131038APending Publication Date: 2026-06-02ZHEJIANG NOYETEC TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ZHEJIANG NOYETEC TECH CO LTD
Filing Date
2026-01-30
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Existing EMI testing systems suffer from rigid testing paths when faced with complex and ever-changing interference scenarios, resulting in low efficiency or insufficient test completeness, and lack the ability to make autonomous decisions and continuously evolve.

Method used

By fusing domain knowledge with real-time acquired spectrum data and test process data, a domain-specific state representation is generated. Combined with a rule engine and an online reinforcement learning model, test sequences are dynamically planned and optimized to generate adaptive test sequences.

Benefits of technology

It has achieved an efficient, complete and standard-compliant EMI testing process, enhancing the system's autonomous decision-making and continuous evolution capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122131038A_ABST
    Figure CN122131038A_ABST
Patent Text Reader

Abstract

This application discloses an automated and adaptive test sequence generation system for EMI testing, belonging to the field of automation technology. By fusing domain knowledge with real-time spectrum data and test progress data, it generates a domain-specific state representation including interference type identifiers and compliance status parameters. Combined with a rule engine and an online reinforcement learning model, it achieves automated and adaptive generation of EMI test sequences. This method effectively solves the problems of low efficiency and insufficient completeness of traditional fixed test sequences in complex interference scenarios. It can understand the test status in real time, dynamically plan and optimize the test path, thereby ensuring that the testing process is compliant with standards, efficient, and comprehensive.
Need to check novelty before this filing date? Find Prior Art