An AI-assisted partial discharge detection and diagnosis method

By employing an AI-assisted partial discharge detection and diagnosis method, and utilizing a cascaded AI model and a multi-dimensional knowledge base, an intelligent end-to-end process from data acquisition to diagnostic results has been achieved. This solves the problem of relying on human experience in existing technologies, and improves detection efficiency and accuracy.

CN122449286APending Publication Date: 2026-07-24GLOBAL SCI & TECH (SHANGHAI) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
GLOBAL SCI & TECH (SHANGHAI) CO LTD
Filing Date
2025-11-26
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Current partial discharge detection relies on engineers' professional experience, lacks real-time interactive intelligent guidance, and lacks self-optimization capabilities, resulting in low detection efficiency and a high false detection rate.

Method used

The AI-assisted partial discharge detection and diagnosis method achieves full-process intelligence from data acquisition to diagnostic results through a first AI model and a second AI model connected in series. Combined with a multi-dimensional knowledge base and a time-sharing reuse mechanism, it provides real-time interactive suggestions and self-optimization.

Benefits of technology

It improves the level of automation and standardization of testing, reduces reliance on expert experience, enhances the accuracy and efficiency of diagnosis, and can provide reliable diagnostic recommendations in complex environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure HDA0005706921330000011
    Figure HDA0005706921330000011
Patent Text Reader

Abstract

The application discloses an AI-assisted partial discharge detection and diagnosis method, comprising the following steps: step S1: in response to a user's on-site detection request, controlling a detection device to collect on-site pulse signals and processing the pulse signals into partial discharge data; step S2: analyzing the partial discharge data based on an AI-assisted diagnosis module and outputting a diagnosis result; and step S3: recording and saving the diagnosis result, the partial discharge data and related parameters as a new detection case and updating to a first database. The AI-assisted diagnosis module is embedded in the whole process from 'initiating a request' to'saving a case', so that not only the data can be analyzed, but also the next operation can be actively recommended and experience can be automatically accumulated, the automation level and the standardization degree of on-site work are improved, and the dependence of the diagnosis process on expert experience is reduced.
Need to check novelty before this filing date? Find Prior Art