Dynamic rotation heliostat mirror surface damage identification method based on unmanned aerial vehicle

By using UAV pose information compensation and a multi-scale attention fusion network, the challenges of image acquisition and damage identification for dynamically rotating heliostats were solved, achieving high-precision and intelligent mirror damage detection, adapting to complex working conditions, and constructing an end-to-end intelligent operation and maintenance closed loop.

CN122454466AActive Publication Date: 2026-07-24XIANGTAN UNIV
5 Cites 0 Cited by

Patent Information

Application Number
CN202610920685.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-25
Publication Date
2026-07-24
Estimated Expiration
2046-06-25

AI Technical Summary

Technical Problem

Existing technologies struggle to achieve efficient and stable image acquisition and micro-damage identification on dynamically rotating heliostats, resulting in problems such as large monitoring blind spots, difficulty in capturing dynamic targets, and inability to track in real time.

Method used

A method for identifying damage to a rotating heliostat mirror based on unmanned aerial vehicles (UAVs) is adopted. Through the technical route of 'motion decoupling-instance tracking-multi-scale fusion', the UAV pose information is used to compensate for image jitter, and a multi-scale attention fusion network is combined to identify damage. This generates a visual data stream with a stable background that preserves the mirror's motion state, and performs pixel-level mirror segmentation and cross-frame identity association.

Benefits of technology

It achieves fully automatic, high-precision, and intelligent sensing and identification of dynamic rotating heliostats, improves the robustness of micro-damage identification, reduces the false detection rate of mirror reflection artifacts, and constructs an end-to-end intelligent operation and maintenance closed loop, adapting to complex environments under different working conditions.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a dynamic rotation heliostat mirror surface damage identification method based on an unmanned aerial vehicle. The core is to construct an intelligent sensing architecture of'air-ground cooperation ', and a video stream of a moving mirror surface is collected in real time through a visual sensor carried by an unmanned aerial vehicle platform; a dynamic image stabilization algorithm based on depth feature optical flow and motion compensation is adopted to counteract unmanned aerial vehicle pose disturbance and target motion coupling interference; meanwhile, a lightweight and attention-guided multi-scale attention fusion network is designed, and pixel-level sensitive and millimeter-level recognition of typical defects such as microcracks, stains and coating falling is achieved; and finally, through a space-time correlation inference engine, outputting a damage quantification map with a geographic information mark and an operation and maintenance decision suggestion. The technical limitation of static or low-speed target detection is broken through, full-automatic, high-frequency and high-reliability health state inspection of a large-scale rotating mirror array is realized, and the operation and maintenance intelligence level and the power generation efficiency guarantee capability of a photo-thermal power station are remarkably improved.
Need to check novelty before this filing date? Find Prior Art