In:
Journal of Analytical Atomic Spectrometry, Royal Society of Chemistry (RSC), Vol. 38, No. 10 ( 2023), p. 2046-2058
Kurzfassung:
Accurately and timely determining the calorific value of coal is essential for optimizing the washing and processing procedures in coal preparation plants and achieving maximum economic benefits. In this study, a rapid coal calorific value analyzer was developed by combining near-infrared spectroscopy (NIRS) and X-ray fluorescence spectroscopy (XRF), and industrial testing and performance evaluation were conducted at a coal preparation plant. This article focuses on the structure, operational process, analysis model, and industrial testing of a rapid NIRS-XRF coal calorific value analyzer. This analyzer consists of six parts: NIRS module, XRF module, sample delivery module, control module, hydrogen production module, and operation software. We proposed a holistic-segmented model quantitative analysis algorithm based on partial least squares regression (PLSR). The analyzer was used to measure the calorific values of four types of product coals, namely, clean coal, middling coal, slime and gangue, from the daily production of the Duanshi coal preparation plant, and compared with the assay results. The test results showed that the root mean square error of prediction (RMSEP), the average absolute error (AAE) and the average relative error (ARE) of the prediction of the coal calorific value by using the holistic-segmented model decreased from 0.65 MJ kg −1 , 0.55 MJ kg −1 and 4.71% of the traditional holistic model to an average of 0.33 MJ kg −1 , 0.28 MJ kg −1 and 2.71%, respectively, a decrease of nearly double. The average standard deviation (SD) also decreased from 0.29 MJ kg −1 to 0.09 MJ kg −1 , which met the Chinese national standard requirement of less than 0.12 MJ kg −1 . Our proposed model significantly improved the accuracy and repeatability of the measurements. Furthermore, the measurement results of this analyzer showed good consistency with traditional chemical analysis results, indicating its potential for widespread application in industries such as coal mining, washing, power generation, and coking, among others. Therefore, the rapid NIRS-XRF coal calorific value analyzer provides a new intelligent means for the clean and efficient utilization of coal.
Materialart:
Online-Ressource
ISSN:
0267-9477
,
1364-5544
Sprache:
Englisch
Verlag:
Royal Society of Chemistry (RSC)
Publikationsdatum:
2023
ZDB Id:
1484654-8
ZDB Id:
54176-X
SSG:
11
