摘要
针对大型储罐罐底声发射检测过程中的声源定位问题,在深入研究三角定位算法和超定定位算法原理的基础上,采用蒙特卡洛模拟方法定量分析比较了两种算法对罐底声发射源定位的性能,主要包括定位计算时间、定位误差分布以及有效定位率等参数。结果表明三角定位算法比超定定位算法快100倍,但在非理想情况下,即存在时间测量误差和声速误差时,三角定位算法对罐底边缘区域的声源定位误差大,而超定定位算法对罐底边缘区域声源的有效定位率低。现场检测过程中应根据实时性、误差范围、计算机性能等要求选择合适的定位算法。
Getting the Acoustic Emission(AE) source location is an important step in AE testing for tank bottom. Passive loca-tion is the only technology that can be used to get the AE source location. It mainly includes triangulation location algorithm and over determined location algorithm. In order to know the two algorithms' location performance including calculation time, posi-tion error and effective targeting rate, a Monte Carlo simulation experiment is implemented. The results demonstrate that triangu- lation location algorithm is about 100 times faster than the over determined location algorithm. However, the measurement er-rors of time difference and acoustic velocity cause different effects on the location results of the two algorithms. The triangula-tion location algorithm produces large errors for the AE sources located at the edge area of tank bottom, while the over deter-mined algorithm results in low effective targeting rate for the AE sources located at the edge area of tank bottom. Therefore, it should select appropriate location algorithm by considering the requirements of calculation time, error range and computer per-formance.
出处
《计算机工程与应用》
CSCD
2013年第17期230-234,共5页
Computer Engineering and Applications
基金
国家重点基础研究发展计划(973)(No.2011cb610505)
中国博士后科学基金(No.2012M510749)
关键词
罐底检测
声发射检测
无源定位
蒙特卡罗模拟
tank bottom testing
Acoustic Emission(AE) testing
passive location
Monte Carlo simulation