Login / Signup

Negative Pressure Waves Based High Resolution Leakage Localization Method Using Piezoceramic Transducers and Multiple Temporal Convolutions.

Guangmin ZhangSiu Chun Michael HoLinsheng HuoJunxiao Zhu
Published in: Sensors (Basel, Switzerland) (2019)
The negative pressure wave (NPW) signals generated by a pipeline leakage often have a long signal duration. When these signals are utilized to compute the leakage position, the long signal duration will result in a large area being considered as leakage area. The localization resolution is low. A novel high-resolution localization algorithm is developed for pipeline leakage detection using piezoceramic transducers in this paper. The proposed algorithm utilizes multiple temporal convolutions to decrease the localization functional values at the points close to the leakage, in order to reduce the range of the leakage area revealed by the proposed algorithm. As a result, the localization resolution is improved. A measured experiment was conducted to study the proposed algorithm. In the experiment, the proposed algorithm was used to monitor a 55.8 m pressurized pipeline with two controllable valves and two Lead Zirconate Titanate (PZT) sensors. With the aid of the piezoceramic sensor, the experimental results show that the proposed algorithm results in a resolution which is better than that of the traditional method.
Keyphrases
  • machine learning
  • deep learning
  • high resolution
  • neural network
  • mass spectrometry
  • coronary artery disease
  • aortic valve
  • data analysis