THE APPLICATION OF WAVELET ANALYSIS AND NEURAL NETWORK FOR WELD-CRACKING METAL MAGNETIC MEMORY SIGNAL RECOGNITION

摘要：

Metal Magnetic Memory (MMM) technique is a new technique which utilizes the distribution of leakage magnetic field of measured object In this paper, a hydraulic pressure experiment is conducted on a AP15L X70 steel pipe. The MMM signal is processed and analyzed with the employment of wavelet analysis for its energy characteristics. The energy characteristics are men adopted as the input of a BP neural network to recognize the weld-cracking. The experiment shows that the employment of wavelet analysis and BP neural network can satisfy the need of weld-cracking recognition.

作者: Zhanjun Feng Weibin Wang Guangwen Liu Song Lin

作者单位: Petrochina Pipeline R&D Center,Langfang,Hebei,China,065000

会议类型: 国际会议

会议名称: The 3rd World Conference on Safety of Oil and Gas Industry(第三届世界石油天然气工业安全会议 WCOGI 2010)

会议地点: 北京

会议语种:英文

页码: 520-523

在线出版日期: 2010-09-27（万方平台首次上网日期，不代表论文的发表时间）