Mfcc- Lstm Framework for Leak Detection and Leak Size Identification in Gas-Liquid Two-Phase Flow Pipelines Based on Acoustic Emission
Title
Mfcc- Lstm Framework for Leak Detection and Leak Size Identification in Gas-Liquid Two-Phase Flow Pipelines Based on Acoustic Emission
Subject
Pipelines
Flow patterns
Leak detection
Two phase flow
Long short-term memory
Acoustic emission testing
Description
Two-phase gas-liquid flows are crucial to the pipeline system. Due to their complicated flow state, existing leak detection techniques are unsuitable for two-phase flow pipelines. To avoid accidents caused by a leak of pipelines, we present a framework for combining the Mel-frequency cepstral coefficient and long short-term memory (MFCC-LSTM) based on acoustic emission (AE). A series of experiments are performed considering 1152 operating conditions, including flow pattern, leak size, direction, and location. In addition, the detection capability of the different features of AE signals combined with LSTM is discussed. The results show that the recognition accuracy of the MFCC-LSTM framework reaches 98.4%. Then, we further perform leak size identification under different flow patterns and found that the MFCC-LSTM framework still exhibits excellent performance. The proposed MFCC-LSTM framework provides a promising solution to identify the leak state and size in two-phase flow pipelines based on the AE technique. 2023, The Authors. All rights reserved.
Creator
Zhang, Zhiyuan
Xu, Changhang
Xie, Jing
Zhang, Yuan
Liu, Pengqian
Liu, Zichen
Date
2023
Type
journalArticle
Identifier
15565068
10.2139/ssrn.4403080
URL
http://dx.doi.org/10.2139/ssrn.4403080
Citation
Zhang, Zhiyuan et al., “Mfcc- Lstm Framework for Leak Detection and Leak Size Identification in Gas-Liquid Two-Phase Flow Pipelines Based on Acoustic Emission,” Lamar University Midstream Center Research, accessed May 18, 2024, https://lumc.omeka.net/items/show/29245.