摘要
针对天然气管道泄漏因泄漏声波信号信噪比(SNR)过低而难于检测的问题,研究了基于Duffing振子的天然气管道泄漏检测方法。该方法将待检测数据输入Duffing振子系统,以振子系统的状态转化实现非周期信号中周期信号的检测。为了更好地提高Duffing振子的检测性能,在Duffing振子设计阶段,以随机共振的有关理论为基础,通过对系统输出信噪比的优化来实现Duffing振子的参数设计。基于实际天然气管道泄漏数据的测试结果表明,所提出方法可在低信噪比(-68dB)的情况下有效检测出泄漏,具有较好的检测性能。
Considering the gas pipeline leak is difficult to detect due to the very low signal to noise ratio （SNR） of the leakgenerated signals, this paper proposes an algorithm for gas pipeline leak detection based on Duffing oscillators. The algorithm makes test data as the input of the Duffing system, and realizes the frequency component detection within the nonperiodical signals by using the state transfer of the Duffing system. In order to improve the detection performance of Duffing oscillators, the design for parameters of the Duffing system is realized by optimization of the output SNR of the system according to the stochastic resonance theory. The test result based on the real leak data of the gas pipeline shows that the proposed method can effectively detect the leakage with a low SNR （ - 68dB） and has the good detection performance.
出处
《高技术通讯》
EI
CAS
CSCD
北大核心
2010年第6期628-631,共4页
Chinese High Technology Letters
基金
国家自然科学基金(60974059
60736026
60721003)
中国博士后科学基金(20090450402)资助项目
作者简介
女，1980年生，博士；研究方向：故障检测；联系人，E-mail：yanghongying@tsinghua．org．cn
