Lightweight smoke area recognition method and system based on improved YOLO for photovoltaic factory

By improving the YOLO model and combining it with a multi-level learning architecture of visual and thermal imaging image streams, the accuracy and real-time performance issues of smoke recognition in photovoltaic plant areas were resolved, enabling accurate and reliable smoke area recognition in complex environments.

CN122289882APending Publication Date: 2026-06-26CHINA POWER CONSTR NEW ENERGY GRP CO LTD GUIZHOU BRANCH +2
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
CHINA POWER CONSTR NEW ENERGY GRP CO LTD GUIZHOU BRANCH
Filing Date
2026-02-02
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

In the complex and vast environment of photovoltaic plant areas, the accuracy, real-time performance and reliability of smoke area identification are low.

Method used

This paper presents a lightweight method for identifying smoke areas in photovoltaic plants based on YOLO. It utilizes a multi-level learning architecture to combine visual and thermal imaging image streams for smoke probability detection, and achieves accurate and reliable smoke area identification by monitoring attention configuration graphs and temporal dependency aggregation.

Benefits of technology

It achieves accurate, reliable, and adaptive real-time smoke identification in photovoltaic plant areas, improving identification accuracy and real-time performance, and adapting to changing weather conditions and complex environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122289882A_ABST
    Figure CN122289882A_ABST
Patent Text Reader

Abstract

This invention discloses a lightweight method and system for identifying smoke areas in photovoltaic (PV) plants based on an improved YOLO framework. It relates to the field of image recognition technology and includes: multi-point monitoring feature mining based on historical smoke detection logs of the PV plant; real-time monitoring of a multi-sensor array controlled by a monitoring attention-configured graph network to obtain visual and thermal image streams of the plant area; introducing an improved YOLO multi-level learning architecture for visual and thermal smoke probability detection; interference tracing correction and time-dependent aggregation of the first and second smoke detection graph networks; and performing smoke area identification based on the reliable smoke detection graph network to obtain a smoke area identification graph network. This invention solves the technical problems of low accuracy, poor real-time performance, and insufficient reliability in smoke area identification in complex and vast PV plant environments, achieving accurate, reliable, and adaptively focused real-time smoke identification in PV plant areas.
Need to check novelty before this filing date? Find Prior Art