A Quality-Hierarchical Temperature Imaging Network for TDLAS Tomography

Jingjing Si; Gengchen Fu; Rui Zhang; Chang Liu

A Quality-Hierarchical Temperature Imaging Network for TDLAS Tomography

Jingjing Si, Gengchen Fu, Rui Zhang, Chang Liu

School of Engineering

Research output: Contribution to conference › Paper › peer-review

Abstract / Description of output

Tunable diode laser absorption spectroscopy (TDLAS) tomography is a well-established technique for combustion diagnostics, which is capable of imaging the 2-D distribution of critical flow-field parameters over cross section of the flame. Reconstruction quality and time resolution are two key aspects that need to be compromised in dynamic monitoring. In this paper, we develop a quality-hierarchical TDLAS tomographic algorithm based on Long Short Term Memory (LSTM) network. From limited amount of line-of- sight projections measured on current tomographic field, this algorithm outputs a coarse-quality temperature image and a fine-quality temperature image with different computational costs. Simulation results validated the efficiency-effectiveness trade-off achieved by this quality-hierarchical temperature imaging network.

Original language	English
Number of pages	5
Publication status	Accepted/In press - 2021
Event	IEEE International Instrumentation and Measurement Technology Conference: I2MTC - Duration: 17 May 2021 → 20 May 2021 https://i2mtc2021.ieee-ims.org/

Conference

Conference	IEEE International Instrumentation and Measurement Technology Conference
Period	17/05/21 → 20/05/21
Internet address	https://i2mtc2021.ieee-ims.org/

Access to Document

qht-lstm-short-final-2021-02-08Accepted author manuscript, 642 KB

1 Article

A Quality-Hierarchical Temperature Imaging Network for TDLAS Tomography
Si, J., Fu, G., Cheng, Y., Zhang, R., Enemali, G. & Liu, C., 18 Jan 2022, In: IEEE Transactions on Instrumentation and Measurement. 71, 4500710.
Research output: Contribution to journal › Article › peer-review

Cite this

@conference{3935895c83514b50b2d7d2239d01416a,

title = "A Quality-Hierarchical Temperature Imaging Network for TDLAS Tomography",

abstract = "Tunable diode laser absorption spectroscopy (TDLAS) tomography is a well-established technique for combustion diagnostics, which is capable of imaging the 2-D distribution of critical flow-field parameters over cross section of the flame. Reconstruction quality and time resolution are two key aspects that need to be compromised in dynamic monitoring. In this paper, we develop a quality-hierarchical TDLAS tomographic algorithm based on Long Short Term Memory (LSTM) network. From limited amount of line-of- sight projections measured on current tomographic field, this algorithm outputs a coarse-quality temperature image and a fine-quality temperature image with different computational costs. Simulation results validated the efficiency-effectiveness trade-off achieved by this quality-hierarchical temperature imaging network.",

author = "Jingjing Si and Gengchen Fu and Rui Zhang and Chang Liu",

year = "2021",

language = "English",

note = "IEEE International Instrumentation and Measurement Technology Conference : I2MTC ; Conference date: 17-05-2021 Through 20-05-2021",

url = "https://i2mtc2021.ieee-ims.org/",

}

TY - CONF

T1 - A Quality-Hierarchical Temperature Imaging Network for TDLAS Tomography

AU - Si, Jingjing

AU - Fu, Gengchen

AU - Zhang, Rui

AU - Liu, Chang

PY - 2021

Y1 - 2021

N2 - Tunable diode laser absorption spectroscopy (TDLAS) tomography is a well-established technique for combustion diagnostics, which is capable of imaging the 2-D distribution of critical flow-field parameters over cross section of the flame. Reconstruction quality and time resolution are two key aspects that need to be compromised in dynamic monitoring. In this paper, we develop a quality-hierarchical TDLAS tomographic algorithm based on Long Short Term Memory (LSTM) network. From limited amount of line-of- sight projections measured on current tomographic field, this algorithm outputs a coarse-quality temperature image and a fine-quality temperature image with different computational costs. Simulation results validated the efficiency-effectiveness trade-off achieved by this quality-hierarchical temperature imaging network.

AB - Tunable diode laser absorption spectroscopy (TDLAS) tomography is a well-established technique for combustion diagnostics, which is capable of imaging the 2-D distribution of critical flow-field parameters over cross section of the flame. Reconstruction quality and time resolution are two key aspects that need to be compromised in dynamic monitoring. In this paper, we develop a quality-hierarchical TDLAS tomographic algorithm based on Long Short Term Memory (LSTM) network. From limited amount of line-of- sight projections measured on current tomographic field, this algorithm outputs a coarse-quality temperature image and a fine-quality temperature image with different computational costs. Simulation results validated the efficiency-effectiveness trade-off achieved by this quality-hierarchical temperature imaging network.

M3 - Paper

T2 - IEEE International Instrumentation and Measurement Technology Conference

Y2 - 17 May 2021 through 20 May 2021

ER -

A Quality-Hierarchical Temperature Imaging Network for TDLAS Tomography

Abstract / Description of output

Conference

Access to Document

Fingerprint

Research output

A Quality-Hierarchical Temperature Imaging Network for TDLAS Tomography

Cite this