A synchronized LIBS-NIRS analysis method for coal properties featured with detection distance correction.

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Materials Science

Background:

• Accurate online coal analysis is crucial for efficient, low-carbon power generation.
• Existing Laser-Induced Breakdown Spectroscopy (LIBS) and Near-Infrared Reflectance Spectroscopy (NIRS) methods lack synchronized detection, hindering practical application.
• Dynamic variations in coal thickness on conveyor belts affect detection distance and spectral quality.

Purpose of the Study:

• To develop a synchronized LIBS and NIRS detection system for real-time coal analysis.
• To propose a spectral correction method to address detection distance variations and improve accuracy.
• To validate the system's performance for online coal property determination.

Main Methods:

• Development of a synchronized LIBS and NIRS detection system with sequence control.
• Implementation of a spectral correction method using NIRS for distance identification and characteristic peak analysis.
• Hybrid spectral reconstruction integrating Principal Component Analysis (PCA), Random Forest (RF), and SHapley Additive exPlanations (SHAP).

Main Results:

• The synchronized system achieved online and real-time analysis of coal properties.
• The spectral correction method effectively mitigated fluctuations caused by varying coal thickness.
• Achieved low average absolute errors for ash (0.405%), moisture (0.143%), calorific value (0.258 MJ/kg), volatile matter (0.117%), fixed carbon (0.546%), and total sulfur (0.036%).

Conclusions:

• The developed synchronized LIBS-NIRS system with spectral correction enables accurate online coal analysis.
• This approach overcomes limitations of previous methods, offering a practical solution for power plants.
• The findings support the use of this integrated spectroscopic technique for real-time coal quality monitoring.