本文已被:浏览次 下载次 |
码上扫一扫! |
|
|
Pipeline acoustic propagation features and recognition of typical damage sounds |
ZHAO Hong-hua, AI Chang-sheng
|
(Department of Mechanical and Electronic Engineering, Jinan University, Jinan 250022, China)
|
Abstract: |
In order to protect pipeline transportation and prevent manmade damages of shemaking hole to steal oil, it is very important to carry out such researches as active protecting and accurate positioning. By analyzing the features of pipelines including wave-guiding and typical acoustic spectrum of manmade damages, a model composed of characters of acoustic wave was established. The results show that acoustic wave in pipeline propagates as plane wave with little falloff. A typical damage acoustic wave is able to propagate for a long distance. The typical acoustic spectrum can be picked up and recognized by the aigorithms of hidden Markov model (HMM) and artifical neural network(ANN), which provides a new method for the prevention of theft in pipeline transportation system. |
Key words: pipeline guided wave damage time-frequency analysis recognition |
|
|