Electrical Resistance Tomography and Flow Pattern Identification Method Based on Deep Residual Neural Network

doi:10.16182/j.issn1004731x.joss.21-0294

Abstract

Abstract:

Aiming at the low accuracy of inverse problem imaging and flow pattern recognition in electrical resistance tomography (ERT), a two-phase flow electrical resistance tomography and flow pattern recognition method based on the deep residual neural network is proposed. The finite element method is used to model the ERT forward problem to construct the "boundary voltage-conductivity distribution-flow pattern category" dataset of various gas-liquid two-phase flow distributions. The residual neural network for ERT image reconstruction and flow pattern identification of gas-liquid two-phase flow is built and trained. The two outputs of the residual neural network are processed respectively to obtain the reconstructed conductivity distribution images and flow pattern identification results. The simulation and static experimental results show that the method can achieve the requirements of imaging and flow pattern identification simultaneously, and has the characteristics of high precision of reconstructed images. And the model is strong generalization and robustness to noise, and has the high accuracy of flow pattern identification.

Key words: residual neural network, deep learning, electrical resistance tomography (ERT), image reconstruction, flow pattern identification

CLC Number:

TP274
TP391

Weiguo Tong, Shichao Zeng, Lifeng Zhang, Zhe Hou, Jiayue Guo. Electrical Resistance Tomography and Flow Pattern Identification Method Based on Deep Residual Neural Network[J]. Journal of System Simulation, 2022, 34(09): 2028-2036.

Figures/Tables 14

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Table 1

Fig. 6

Table 2

Fig. 7

Fig. 8

Table 3

Table 4

Fig. 9

Fig. 10

References 17

[1]	陶芳芳, 宁尚雷, 靳海波. 电阻层析成像技术在气液(固)多相流动体系中的应用进展[J]. 过程工程学报, 2020, 20(4): 371-381.
	Tao Fangfang, Ning Shanglei, Jin Haibo. Application Progress of Electrical Resistance Tomography in Gas-Liquid (Solid) Multiphase Flow Systems[J]. The Chinese Journal of Process Engineering, 2020, 20(4): 371-381.
[2]	陈宇, 夏宗基, 周雨佳. 基于修正稀疏拟牛顿的电容层析成像重建算法[J]. 系统仿真学报, 2019, 31(5): 819-827.
	Chen Yu, Xia Zongji, Zhou Yujia. Electrical Capacitance Tomography Reconstruction Algorithm Based on Modified Sparse Quasi-Newton[J]. Journal of System Simulation, 2019, 31(5): 819-827.
[3]	肖理庆, 王化祥. 基于聚类电阻层析成像静态图像重建算法[J]. 仪器仪表学报, 2016, 37(6): 1258-1266.
	Xiao Liqing, Wang Huaxiang. Static Image Reconstruction Algorithm Based on Clustering Electrical Resistance Tomography[J]. Chinese Journal of Scientific Instrument, 2016, 37(6): 1258-1266.
[4]	De Kerret F, Béguin C, Etienne S. Two-Phase Flow Pattern Identification in a Tube Bundle Based on Void Fraction and Pressure Measurements with Emphasis on Churn Flow[J]. International Journal of Multiphase Flow (S0301-9322), 2017, 94: 94-106.
[5]	李凯锋, 王保良, 黄志尧, 等. K-均值聚类在CCERT系统流型辨识中的应用[J]. 北京航空航天大学学报, 2017, 43(11): 2280-2285.
	Li Kaifeng, Wang Baoliang, Huang Zhiyao, et al. Application of K-means Clustering in Flow Pattern Identification of CCERT System[J]. Journal of Beijing University of Aeronautics and Astronautics, 2017, 43(11): 2280-2285.
[6]	吴新杰, 黄国兴, 王静文. 压缩感知在电容层析成像流型辨识中的应用[J]. 光学精密工程, 2013, 21(4): 1062-1068.
	Wu Xinjie, Huang Guoxing, Wang Jingwen. Application of Compressed Sensing in Flow Pattern Identification of Electrical Capacitance Tomography[J]. Optics and Precision Engineering, 2013, 21(4): 1062-1068.
[7]	叶明, 李晓丞, 刘凯, 等. 一种基于U²-Net模型的电阻抗成像方法[J]. 仪器仪表学报, 2021, 42(2): 235-243.
	Ye Ming, Li Xiaocheng, Liu Kai, et al. An Electrical Impedance Imaging Method Based on U²-Net Model[J].Chinese Journal of Scientific Instrument, 2021, 42(2): 235-243.
[8]	Li F, Tan C, Dong F. Electrical Resistance Tomography Image Reconstruction with Densely Connected Convolutional Neural Network[J]. IEEE Transactions on Instrumentation and Measurement (S0018-9456), 2021, 70: 1-11.
[9]	Wu Y, Chen B, Liu K, et al. Shape Reconstruction with Multiphase Conductivity for Electrical Impedance Tomography Using Improved Convolutional Neural Network Method[J]. IEEE Sensors Journal (S1530-437X), 2021, 21(7): 9277-9287.
[10]	Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-scale Image Recognition[C]//International Conference on Learning Representations.San Diego, CA, USA: ICLR, 2015: 1-14.
[11]	宋蕾, 陈德运, 姚玉梅, 等. Elman神经网络在ECT系统流型辨识中的应用[J]. 哈尔滨理工大学学报, 2014, 19(5): 103-108.
	Song Lei, Chen Deyun, Yao Yumei, et al. Application of Elman Neural Network in ECT System Flow Pattern Identification[J]. Journal of Harbin University of Science and Technology, 2014, 19(5): 103-108.
[12]	李岩, 王璐, 李佳琪. 基于改进ALEXNET卷积神经网络的电容层析成像三维图像重建[J]. 哈尔滨理工大学学报, 2020, 25(4): 109-115.
	Li Yan, Wang Lu, Li Jiaqi. Three-dimensional Image Reconstruction of Electrical Capacitance Tomography Based on Improved ALEXNET Convolutional Neural Network[J]. Journal of Harbin University of Science and Technology, 2020, 25(4): 109-115.
[13]	李峰, 谭超, 董峰. 全连接深度网络的电学层析成像算法[J]. 工程热物理学报, 2019, 40(7): 1526-1531.
	Li Feng, Tan Chao, Dong Feng. Electrical Tomography Algorithm Based on Fully Connected Deep Network[J].Journal of Engineering Thermophysics, 2019, 40(7): 1526-1531.
[14]	肖理庆. 电阻层析成像有限元模型优化与图像重建算法研究[D]. 天津: 天津大学, 2014.
	Xiao Liqing. Research on Finite Element Model Optimization and Image Reconstruction Algorithm of Electrical Resistance Tomography[D]. Tianjin: Tianjin University, 2014.
[15]	张立峰, 王化祥. 一种修正的电阻层析成像Landweber迭代算法[J]. 计量学报, 2016, 37(3): 271-274.
	Zhang Lifeng, Wang Huaxiang. A Modified Landweber Iterative Algorithm for Electrical Resistance Tomography[J]. Acta Metrology Sinica, 2016, 37(3): 271-274.
[16]	徐胜军, 欧阳朴衍, 郭学源, 等. 多尺度特征融合空洞卷积ResNet遥感图像建筑物分割[J]. 光学精密工程, 2020, 28(7): 1588-1599.
	Xu Shengjun, Ouyang Puyan, Guo Xueyuan, et al. Multiscale Feature Fusion Cavity Convolution ResNet Remote Sensing Image Building Segmentation[J]. Optics and Precision Engineering, 2020, 28(7): 1588-1599.
[17]	He K, Zhang X, Ren S, et al. Deep Residual Learning for Image Recognition[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA: IEEE, 2016: 770-778.

算法	ICC	RIE
Landweber	0.608	0.801
FCDN	0.927	0.319
Vgg16	0.967	0.219
h-Net	0.972	0.202

算法	ICC	RIE
Landweber	0.526	1.249
FCDN	0.835	0.504
Vgg16	0.793	0.533
h-Net	0.894	0.379

流型	样本数	正确数	Acc/%
环流	480	479	99.8
单泡流	495	495	100
双泡流	509	509	100
三泡流	511	511	100
四泡流	505	505	100