采用多任务学习和电容层析成像的两相流参数测量方法

杨海潮; 胡红利; 陆程程; 卢家宇; 王茂杰

doi:10.7652/xjtuxb202303019

您当前的位置：

首页 >

文章列表页 >

采用多任务学习和电容层析成像的两相流参数测量方法

—————————————————————— | 更新时间：2025-07-09

- 采用多任务学习和电容层析成像的两相流参数测量方法
- Two-Phase Flow Parameter Measurement Using Multi-Task Learning and Electrical Capacitance Tomography
- 西安交通大学学报 2023年57卷第3期页码：202-211
- 作者机构：
  
  西安交通大学电力设备电气绝缘国家重点实验室,西安,710049
- 作者简介：
  
  : 2022-09-28。作者简介: 杨海潮(1999—),男,硕士生
  胡红利(通信作者),男,教授,博士生导师。基金项目: 国家自然科学基金资助项目(52177009)
- 基金信息：
  
  : 2022-09-28。作者简介: 杨海潮(1999—),男,硕士生;胡红利(通信作者),男,教授,博士生导师。基金项目: 国家自然科学基金资助项目(52177009);陕西省科技攻关计划资助项目(2016GY-001)。
- DOI：10.7652/xjtuxb202303019
  中图分类号： TP183
- 网络首发：2023-03-10，
  
  纸质出版：2023
- 稿件说明：
移动端阅览
杨海潮, 胡红利, 陆程程, 等. 采用多任务学习和电容层析成像的两相流参数测量方法[J]. 西安交通大学学报, 2023,57(3):202-211.

YANG Haichao, HU Hongli, LU Chengcheng, et al. Two-Phase Flow Parameter Measurement Using Multi-Task Learning and Electrical Capacitance Tomography[J]. 2023, 57(3): 202-211.
杨海潮, 胡红利, 陆程程, 等. 采用多任务学习和电容层析成像的两相流参数测量方法[J]. 西安交通大学学报, 2023,57(3):202-211. DOI： 10.7652/xjtuxb202303019.

YANG Haichao, HU Hongli, LU Chengcheng, et al. Two-Phase Flow Parameter Measurement Using Multi-Task Learning and Electrical Capacitance Tomography[J]. 2023, 57(3): 202-211. DOI： 10.7652/xjtuxb202303019.

摘要

针对电容层析成像技术(ECT)重建的图像精度低、对流动参数测量误差较大的问题

提出了一种采用多任务学习和电容层析成像的两相流参数测量方法。首先

采用残差模块作为特征提取模块对Landweber算法重构的图像进行特征提取

以此消除重建图像的伪影对流动参数测量所带来的干扰; 其次

为了进一步提升模型的抗噪声干扰能力

采用硬参数共享的双路输出神经网络结构进行初次相含率测量; 之后

采用融合神经网络与极端随机森林等分类模型的方法来实现流型识别; 最后

为了进一步提升相含率测量的准确度

采用极端随机森林等回归模型通过融合流型识别和初次相含率测量的结果

获得最终的相含率。实验结果表明:该方法相对于仅采用神经网络的方法

可以提升相含率测量的准确度; 在实测数据测试中

该方法进行相含率测量的相对误差在5%以内

流型识别的准确度可以达到98%以上

表现出了较好的抗噪声干扰能力

能够满足对实际工业应用的需求。

Abstract

A two-phase flow parameter measurement method

which combines multi-task learning and electrical capacitance tomography

is proposed to solve the problem of the interference in parameter measurement caused by the low accuracy of the images reconstructed by the reconstruction algorithm in terms of electrical capacitance tomography. Firstly

the residual module is used as the feature extraction module to extract the features of the images reconstructed by the Landweber algorithm

so as to eliminate the interference of artifacts of reconstructed images on parameter measurement. Secondly

a dual output multi-task neural network with hard parameter sharing is built to finish the task of initial measurement of the phase fraction in order to further improve the anti-noise ability of the model. Then

a method of fusing neural network and classification models

such as extreme random forests

is proposed to achieve flow identification. Finally

regression models such as extreme random forests are used to obtain the final phase fraction by fusing the results of flow pattern identification and initial results of the phase fraction in order to further improve the accuracy of phase fraction measurement. The experimental results show that the proposed algorithm can improve the accuracy of phase fraction measurement when compared with the method of only using the neural network. In the actual measurement

the relative error of phase fraction measured by the proposed algorithm is within 5%

and the accuracy of flow identification is beyond 98%

which shows a good anti-noise ability of the model and can meet the demands of practical industrial applications.

关键词

Keywords

references

WANTA D, MAKOWIECKA O, SMOLIK W T, et al. Numerical evaluation of complex capacitance measurement using pulse excitation in electrical capacitance tomography [J]. Electronics, 2022, 11(12): 1864.

RASEL R K, STRAITON B, MARASHDEH Q, et al. Toward water volume fraction calculation in multiphase flows using electrical capacitance tomography sensors [J]. IEEE Sensors Journal, 2021, 21(6): 7702-7712.

SUN Yurong, ZHANG Yuyan, WEN Yintang. Identification of defects in the inner layers of composite components based on capacitive sensing [J]. Review of Scientific Instruments, 2022, 93(9): 094710.

张立峰, 苗雨. 基于电容层析成像系统测量信号稀疏性的两相流流型辨识 [J]. 计量学报, 2021, 42(7): 861-865.

ZHANG Lifeng, MIAO Yu. Identification of two-phase flow pattern based on the sparsity of measured capacitance for electrical capacitance tomography [J].Acta Metrologica Sinica, 2021, 42(7): 861-865.

WAJMAN R, BANASIAK R, BABOUT L. On the use of a rotatable ECT sensor to investigate dense phase flow: a feasibility study [J]. Sensors, 2020, 20(17): 4854.

ZHU Hai, SUN Jiangtao, LONG Jun, et al. Deep image refinement method by hybrid training with images of varied quality in electrical capacitance tomography [J]. IEEE Sensors Journal, 2021, 21(5): 6342-6355.

RODRIGUEZ-FRIAS M A, YANG Wuqiang. Dual-modality 4-terminal electrical capacitance and resistance tomography for multiphase flow monitoring [J]. IEEE Sensors Journal, 2020, 20(6): 3217-3225.

SHEN Jingjing, MENG Shuanghe, WANG Jing, et al. Study on the shape of staggered electrodes for 3-D electrical capacitance tomography sensors [J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-10.

HOSSAIN M S, ABIR M T, ALAM M S, et al. An algorithm to image individual phase fractions of multiphase flows using electrical capacitance tomography [J]. IEEE Sensors Journal, 2020, 20(24): 14924-14931.

LIU Kunhua, YE Zihao, GUO Hongyan, et al. FISS GAN: a generative adversarial network for foggy image semantic segmentation [J]. IEEE/CAA Journal of Automatica Sinica, 2021, 8(8): 1428-1439.

MOSTOFA M, FERDOUS S N, RIGGAN B S, et al. Joint-SRVDNet: joint super resolution and vehicle detection network [J]. IEEE Access, 2020, 8: 82306-82319.

HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ, USA: IEEE, 2016: 770-778.

SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition [C/OL]//Proceedings of the 3rd International Conference on Learning Representations. London, UK: ICLR, 2015[2022-06-20]. https://www.engineeringvillage.com/app/doc/?docid=cpx_M2fa5a91416ccf2afed5M7ae 310178163131pageSize=25index=1searchId=11caf3da2de742aca9c94f85991467f2resultsCount=1 usageZone=resultslistusageOrigin=searchresults searchType=Quick.

HUANG Gao, LIU Zhuang, VAN DER MAATEN L, et al. Densely connected convolutional networks [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ, USA: IEEE, 2017: 2261-2269.

ZHAO Jiejie, DU Bowen, SUN Leilei, et al. Deep multi-task learning with relational attention for business success prediction [J]. Pattern Recognition, 2021, 110: 107469.

PATEL V S, NIE Zhongliang, LE T N, et al. Masked face analysis via multi-task deep learning [J]. Journal of Imaging, 2021, 7(10): 204.

ZHANG Yu, YANG Qiang. A survey on multi-task learning [J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(12): 5586-5609.

王化祥, 王超, 陈磊. 基于Landweber迭代的图像重建算法 [J]. 信号处理, 2000, 16(4): 354-356, 328.

WANG Huaxiang, WANG Chao, CHEN Lei. An image reconstruction algorithm based on the Landweber iteration method [J]. Signal Processing, 2000, 16(4): 354-356, 328.

YE Jiamin, YANG Wuqiang, WANG Chao. Investigation of spatial resolution of electrical capacitance tomography based on coupling simulation [J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(11): 8919-8929.

ZHENG Jin, PENG Lihui. A deep learning compensated back projection for image reconstruction of electrical capacitance tomography [J]. IEEE Sensors Journal, 2020, 20(9): 4879-4890.

ZHU Hai, SUN Jiangtao, XU Lijun, et al. Permittivity reconstruction in electrical capacitance tomography based on visual representation of deep neural network [J]. IEEE Sensors Journal, 2020, 20(9): 4803-4815.

ZHANG Lifeng, DAI Li. Image reconstruction of electrical capacitance tomography based on adaptive support driven Bayesian reweighted algorithm [J]. IEEE Sensors Journal, 2021, 21(18): 20648-20656.

OSPINA-ACERO D, MARASHDEH Q M, TEIXEIRA F L. Relevance vector machine image reconstruction algorithm for electrical capacitance tomography with explicit uncertainty estimates [J]. IEEE Sensors Journal, 2020, 20(9): 4925-4939.

LI Jianwei, YANG Xiaoguang, WANG Youhua, et al. An image reconstruction algorithm based on RBF neural network for electrical capacitance tomography [C]//2012 Sixth International Conference on Electromagnetic Field Problems and Applications. Piscataway, NJ, USA: IEEE, 2012: 1-4.

陈露阳, 尹佳雯, 孙志强, 等. 基于EEMD-Hilbert谱的气液两相流钝体绕流流型识别 [J]. 仪器仪表学报, 2017, 38(10): 2536-2546.

CHEN Luyang, YIN Jiawen, SUN Zhiqiang, et al.Flow regime identification of gas-liquid two-phase flow with flow around bluff-body based on EEMD-Hilbert spectrum [J]. Chinese Journal of Scientific Instrument, 2017, 38(10): 2536-2546.

DEABES W, ABDEL-HAKIM A E, BOUAZZA K E, et al. Adversarial resolution enhancement for electrical capacitance tomography image reconstruction [J]. Sensors, 2022, 22(9): 3142.

OTSU N. A threshold selection method from gray-level histograms [J]. IEEE Transactions on Systems, Man, and Cybernetics, 1979, 9(1): 62-66.

王小鑫. 基于静电/电容传感器的气固两相流动参数检测方法研究 [D]. 西安: 西安交通大学, 2017.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用曲波变换和电容层析成像的两相流含水率测量方法

不同跨临界CO2直冷系统架构下动力电池冷却系统性能分析

基于VOF模型的相界面面积和形心计算方法

弦月型通道热虹吸提升管流动特性的数值计算