An intelligent detection and identification device and method for a power battery liquid cooling radiator

By combining a non-contact thermal excitation source with a multi-source synchronous sensing module and utilizing a four-dimensional tensor construction and registration module, the problem of misjudging minute surface defects and locating hidden internal defects in liquid-cooled radiators was solved, achieving high-speed and accurate defect identification and location, and improving the robustness of the detection system.

CN122409753APending Publication Date: 2026-07-17广西易德科技有限责任公司
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
广西易德科技有限责任公司
Filing Date
2026-05-19
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

In existing technologies, liquid-cooled radiators have high false detection rates for surface micro-defects, difficulty in locating internal hidden defects, and low detection efficiency. Multi-source heterogeneous data cannot achieve cross-modal cross-judgment, making it difficult to identify deep structural damage.

Method used

By combining a non-contact transient thermal excitation source with a multi-source synchronous sensing module, and through a four-dimensional tensor construction and registration module, the synchronous acquisition and fusion judgment of surface and internal features are realized. The internal flow channel state is inverted using the thermal time constant, and the adaptive anti-reflective fusion mechanism reduces misjudgment, thus achieving high-speed and accurate defect identification.

Benefits of technology

It achieves high-speed, non-contact, precise localization of internal hidden defects, reduces false alarm rate, improves the robustness of the detection system, shortens detection time to the second level, and accurately outputs the three-dimensional coordinates of internal blockage points.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122409753A_ABST
    Figure CN122409753A_ABST
Patent Text Reader

Abstract

本发明公开一种动力电池液冷散热器智能检测识别装置及方法,涉及无损检测与机器视觉领域。针对铝合金高反光致表面缺陷误判及传统流体测试慢、无法精准定位内部堵塞等问题,本装置包括多源同步感知模组、高速时序同步控制及边缘计算单元。检测时,依据位置脉冲触发微秒级飞拍;将多源异构数据逆向投影构建四维特征张量,解算瞬态温度衰减的热时间常数以反演内部堵塞状态;同时依据灰度饱和度生成抑制权重,利用3D深度特征对2D高反光区进行加权掩码补偿,最终执行联合交叉判定。本发明实现了内外部缺陷的联合三维精准定位,大幅提升了检测鲁棒性。
Need to check novelty before this filing date? Find Prior Art