Purposes In the real-time monitoring process of medium-to-low ash content in coal, there exist problems such as low monitoring accuracy, long detection cycle, and high radiation risk. To address these issues, in this study, a detection model and corresponding measurement method are proposed on the basis of a combination of dual-energy X-ray measurement results and artificial intelligence algorithms, namely the coal ash content detection model based on the Multilayer Perceptron neural network. Methods First, a low-ash coal ash content detection experimental system was designed on the basis of dual-energy X-ray technology. By considering variables such as particle size, X-ray incident angle, humidity, and composition, the effects of different particle sizes, X-ray incident angles, humidity levels, and different components on the identification of ash content in low-ash coal were investigated. Subsequently, an accurate mapping relationship between X-ray transmission data and sample ash content was established. Finally, the neural network was trained by using dual-energy X-ray data to ensure the accuracy and reliability of the model predictions. Results The results show that the root means square error between the predicted values and the true values after training the model is 0.081 7, and the absolute value of the average error is 0.000 01. When the ash content is less than 20%, the range deviation of the predictions is ±0.79%; for ash content ranging from 20% to 35%, the range deviation is ±0.94%; for ash content above 35%, the range deviation is larger.
QIY M.The application status and research of on-line coal quality analysis technology[J].Urban and Rural Studies,2021(6):386-387.
[3]
赵忠辉,方全国.煤质在线检测技术现状及发展趋势分析[J].煤质技术,2017(4):18-21.
[4]
ZHAOZ H, FANGQ G.Present situation and development trend analysis of coal quality on-line detection technology[J].Goal Quality Technology,2017(4):18-21.
[5]
KONGL X, BAIJ, LIH Z,et al.The mineral evolution during coal washing and its effect on ash fusion characteristics of Shanxi high ash coals[J].Fuel,2018,212:268-273.
[6]
LIUY P, ZHONGB J, LIS H.Study on impact of coal particle ash content on bottom ash formation behavior in fluidized bed combustion[C]//Industrial Electronics & Applications.IEEE,2014.
[7]
YUANX F, WUY T, SUNL H,et al.Research on efficient construction paths for intelligent coal mines in china from the configuration perspective[J].Applied Sciences,2023,13(1):673.
[8]
张宏亮,林木松.煤质快速分析仪器应用现状[J].热力发电,2002(4):7-9.
[9]
ZHANGH L, LINM S.Present situation of using coal quality expressanalysers[J].Thermal Power Generation,2002(4):7-9.
[10]
钱坤,陈开玲.煤炭灰分在线检测技术对比分析[J].煤炭加工与综合利用,2020(3):62-65.
[11]
QIANK, CHENK L.Comparative analysis of online detection technology for coal ash content[J].Coal Processing & Comprehensive Utilization,2020(3):62-65.
[12]
RIZKR A M, EL-KATEBA H, ABDUL-KADERA M.On-line nuclear ash gauge for coal based on gamma-ray transmission techniques[J].Journal of radioanalytical and nuclear chemistry,1999,242(1):139-145.
CHENGD, TANGX D, LIF H,et al.Research on method to optimize accuracy of γ-ray backscatter ash analyzer[J].Journal of Hunan University(Natural Sciences),2019,46(2):92-96.
[15]
张泽琳,杨建国.煤灰分在线检测方法及设备[J].选煤技术,2012(2):59-63,72.
[16]
ZHANGZ L, YANGJ G.Research on process and equipment of coal ash content on- line detection[J].Goal Preparation Technology,2012(2):59-63,72.
LI.C Y,CHI F,LIU C R.Application of partial least squares method based on spectral line screening in LIBS-based rapid detection of coal ash[J].Coal Quality Technology,2019(1):37-39.
ZHAOHui.On-line coal ash detection technology of application status and development prospects[J].Coal Processing & Comprehensive Utilization,2022(4):79-82.
PEIW W, MAX M, FANY P,et al.Effect of flotation preparation of ultra-low ash coal based on BPMA prediction[J].Journal of China University of Mining & Technology,2023,52(2):389-396.