Terahertz Coal Ash Prediction Method Based on Dual-Channel Convolutional Neural Network

Jiaojiao Ren; Tiexin Jiao; Jian Gu; Qi Chen; Lijuan Li; Jiyang Zhang

doi:10.3788/AOS231086

Acta Optica Sinica, Volume. 43, Issue 22, 2230001(2023)

Terahertz Coal Ash Prediction Method Based on Dual-Channel Convolutional Neural Network

Jiaojiao Ren^1,2,3、*, Tiexin Jiao^1,2,3, Jian Gu^1,2,3, Qi Chen³, Lijuan Li^1,2,3, and Jiyang Zhang³

¹Key Laboratory of Photoelectric Measurement and Optical Information Transmission Technology of Ministry of Education, Changchun University of Science and Technology, Changchun 130022, Jilin , China

²College of Optoelectronic Engineering, Changchun University of Science and Technology, Changchun 130022, Jilin , China

³Zhongshan Institute of Changchun University of Science and Technology, Zhongshan 528400, Guangdong , China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(9)

$Sample ash mass fraction and thickness$

Fig. 1. Sample ash mass fraction and thickness

Download full size

Fig. 2. Spectrogram of THz spectrometer

Download full size

$Refractive index and absorption coefficient of 10 samples with different ash mass fraction. (a) Refractive index; (b) absorption coefficient$

Fig. 3. Refractive index and absorption coefficient of 10 samples with different ash mass fraction. (a) Refractive index; (b) absorption coefficient

Download full size

Fig. 4. Absorption spectra of nine samples. (a) Original absorption spectra; (b) absorption spectra after thickness corrected enhancement

Download full size

Fig. 5. Structure of dual-channel convolutional neural network

Download full size

Fig. 6. Loss function value in the process of model training

Download full size

Fig. 7. Influence of fusion weight on the accuracy of the algorithm

Download full size

Fig. 8. Test set prediction results

Download full size

Table 1. Prediction results of ash mass fraction under different models

View table

Table 1. Prediction results of ash mass fraction under different models

Model	Training set（N_sam=32）		Prediction set（N_sam=14）
Model	R² /%	$E_{R M S}$	R² /%	$E_{R M S}$
Proposed	98.21	0.1442	93.56	0.2037
CNN	95.38	0.1858	91.33	0.3145
PLSR	91.56	0.2767	87.33	0.4736
BP	89.31	0.4349	85.47	0.5989
LSSVM	90.78	0.2734	86.68	0.4595

Tools

Get Citation

Copy Citation Text

Jiaojiao Ren, Tiexin Jiao, Jian Gu, Qi Chen, Lijuan Li, Jiyang Zhang. Terahertz Coal Ash Prediction Method Based on Dual-Channel Convolutional Neural Network[J]. Acta Optica Sinica, 2023, 43(22): 2230001

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Spectroscopy

Received: Jun. 2, 2023

Accepted: Aug. 3, 2023

Published Online: Nov. 20, 2023

The Author Email: Ren Jiaojiao (zimengrenjiao@163.com)

DOI:10.3788/AOS231086

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology