化工进展 ›› 2024, Vol. 43 ›› Issue (2): 781-790.DOI: 10.16085/j.issn.1000-6613.2023-1383

• 专栏:多相流测试 • 上一篇    下一篇

基于流动噪声解耦的水平管弹状流气弹特性参数测量

李新龙1,2,3(), 张昭1,2,3, 王嘉辉1,2,3, 张延胜1,2,3, 董芳1,2,3()   

  1. 1.河北大学质量技术监督学院,河北 保定 071002
    2.计量仪器与系统国家地方联合工程研究中心,河北 保定 071002
    3.河北省能源计量与安全检测技术重点实验室,河北 保定 071002
  • 收稿日期:2023-08-11 修回日期:2023-10-12 出版日期:2024-02-25 发布日期:2024-03-07
  • 通讯作者: 董芳
  • 作者简介:李新龙(2000—),男,硕士研究生,研究方向为多相流检测技术。E-mail:lxl1587618582@163.com
  • 基金资助:
    国家自然科学基金(62173122);河北省自然科学基金(F2022201034)

Characteristic parameters measurement based on flow noise decoupling in horizontal slug flow

LI Xinlong1,2,3(), ZHANG Zhao1,2,3, WANG Jiahui1,2,3, ZHANG Yansheng1,2,3, DONG Fang1,2,3()   

  1. 1.School of Quality and Technical Supervision, Hebei University, Baoding 071002, Hebei, China
    2.National & Local Joint Engineering Research Center of Metrology Instrument and System, Hebei University, Baoding 071002, Hebei, China
    3.Hebei Key Laboratory of Energy Metering and Safety Testing Technology, Hebei University, Baoding 071002, Hebei, China
  • Received:2023-08-11 Revised:2023-10-12 Online:2024-02-25 Published:2024-03-07
  • Contact: DONG Fang

摘要:

弹状流是一种典型的气液两相间歇流动形态,气弹特性参数的测量对于弹状流摩擦压降和截面含气率等流动参数测量具有重要意义。本文设计了基于声发射原理的气弹参数测量传感器,并对不同流动条件下的水平管弹状流噪声信号进行了测量。利用鲸鱼优化算法(WOA)变分模态分解(VMD)对噪声信号进行处理,结合原始信号从能量和熵的角度进行分析,完成了对水平管弹状流噪声信号的解耦。将噪声信号分解为高频的气固噪声、中频的气液噪声与低频的液固噪声。通过对声发射时域信号分析实现了弹状流频率与气弹长度参数的测量。利用CatBoost算法选取8个特征值构建了弹状流气弹频率和气弹长度的预测模型。实验结果表明,弹状流频率预测模型平均绝对百分比误差(MAPE)为5.12%,95.25%实验点的相对偏差都在±15%范围内,气弹长度预测模型MAPE为7.77%,90.48%实验点的相对偏差都在±15%范围内。

关键词: 弹频, 气弹长度, 声发射, 流动噪声, 能量,

Abstract:

Slug flow is a typical two-phase flow pattern of gas-liquid intermittent flow. The measurement of gas slug characteristic parameters is of great significance for the measurement of abrasion pressure drop in slug flow, sectional void fraction, and other flow parameters. In this paper, a gas slug parameter measurement sensor based on the acoustic emission principle was designed, and the noise signals of horizontal pipe slug flow under different flow conditions were measured. Whale optimization algorithm (WOA) was employed to optimize the variational modal decomposition (VMD) for noise signal processing. By analyzing the original signal from the perspectives of energy and entropy, the decoupling of noise signals from the horizontal pipe slug flow was achieved. The noise signals were divided into high-frequency gas-solid noise, medium-frequency gas-liquid noise, and low-frequency liquid-solid noise. The measurement of gas slug frequency and slug length was realized by analyzing the time-domain signals of acoustic emission. The CatBoost algorithm was used to select 8 feature values to construct predictive models for slug flow slug frequency and slug length. The mean absolute percentage error (MAPE) of the slug frequency prediction model was 5.12%, and the relative deviations of 95.25% of experimental points were within the range of ±15%. The mean absolute percentage error (MAPE) of the slug length prediction model was 7.77%, and the relative deviations of 90.48% of experimental points were within the range of ±15%.

Key words: slug frequency, slug length, acoustic emission, flow noise, energy, entropy

中图分类号: 

京ICP备12046843号-2;京公网安备 11010102001994号
版权所有 © 《化工进展》编辑部
地址:北京市东城区青年湖南街13号 邮编:100011
电子信箱:hgjz@cip.com.cn
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn