Bolt quality testing research using weighted fusion algorithm based on correlation function. (2018)

• Record Type: Journal Article
• Authors: Sun, Xiaoyun; Xing, Hui; Han, Guang; Cheng, Jiulong; Yuan, Yongbang; Bian, Jianpeng; Zheng, Haiqing; Wang, Mingming
• Is Part Of: International journal of modelling, identification and control. Volume 30:Number 1(2018)
• Page Start: 19
• Page End: 29
• Publication Date: 2018
• Subjects: anchor bolt modelling -- non-destructive testing -- D-S fusion algorithm -- weighted fusion based on correlation function; Engineering -- Methodology -- Periodicals; Science -- Methodology -- Periodicals; 001.42
• Languages: English
• ISSNs: 1746-6172

Abstract

Bolt length is an important factor for quality evaluation of anchors. Because of the harsh detection environment and the interference caused by instruments, bolt testing signal contains a lot of noises that make it difficult to analyse and predict the parameters of anchor bolts accurately. In this paper, a non-destructive method based on information fusion of pseudo random signals is presented for bolt quality testing. After pseudo-random signals are generated by multiple sources, weighted fusion algorithm based on correlation function is proposed for fusion processing. Compared with the D-S fusion algorithm and average weighted fusion approach, the correlation fusion is verified to have higher accuracy and better retention of frequency characteristic. Finally, this proposed approach is proved to be more suitable for random signal fusion.