Multi-frequency Fused Lock-in Thermography in Detecting Defects at Different Depths

被引:9
作者
Wei, Yanjie [1 ]
Ye, Yimin [1 ]
He, Hongjun [2 ]
Su, Zhilong [1 ]
Ding, Li [1 ,3 ]
Zhang, Dongsheng [1 ,3 ]
机构
[1] Shanghai Univ, Sch Mech & Engn Sci, Shanghai Inst Appl Math & Mech, Shanghai Key Lab Mech Energy Engn, Shanghai 200444, Peoples R China
[2] Shanghai Acad Spaceflight Technol, Shanghai 201100, Peoples R China
[3] Shanghai Univ, Sch Mech & Engn Sci, 99 Shangda Rd, Shanghai 200444, Peoples R China
基金
中国国家自然科学基金; 中国博士后科学基金; 国家重点研发计划;
关键词
Lock-in thermography; Multi-frequency; Phase images; Non-destructive testing; IMAGE FUSION; PULSED THERMOGRAPHY; COMPOSITE; DAMAGE; TRANSFORM;
D O I
10.1007/s10921-022-00889-8
中图分类号
TB3 [工程材料学];
学科分类号
0805 ; 080502 ;
摘要
Lock-in thermography (LT) is a non-destructive evaluation method capable of inspecting defects in composites by calculating differences of the thermal amplitude and phase between sound and defective regions. This method greatly reduces influences of non-uniform heating and emissivity in detection, while a single excitation frequency limits the depth resolution of the subsurface defects, as defects at a specific depth can only be detected with a specific excitation frequency. In this paper, a multi-frequency fused LT is proposed to improve the depth resolution and obtain better inspection results. Sequential infrared images are acquired under multiple excitation frequencies and then the phase information is extracted along with a specified enhancement scheme. Finally, a fusion image containing defects with multiple depths is constructed by using principal component analysis method. The proposed method has been applied to characterization of the flat-bottomed holes in a carbon fiber reinforced polymer (CFRP) panel and compared with other active thermography techniques. To improve the detection efficiency, an optimal selection method of excitation frequency is also proposed. The experimental results show that the use of the proposed method is able to identify all defects at multiple depths with clear edges and high contrast.
引用
收藏
页数:10
相关论文
empty
未找到相关数据