1.山西天地王坡煤业有限公司,山西 晋城 048000
2.天地(常州)自动化有限公司,江苏 常州 213125
3.中国矿业大学 信息与控制工程学院,江苏 徐州 221116
雷萌,博士,副教授,研究方向:近红外光谱智能识别,E-mail:lmsiee@cumt.edu.cn
纸质出版日期:2024-04-15,
收稿日期:2023-11-26,
修回日期:2023-12-21,
扫 描 看 全 文
王亚栋,贾俊伟,谭韦君等.基于深度残差网络和近红外光谱的煤矸石智能识别[J].分析测试学报,2024,43(04):607-613.
WANG Ya-dong,JIA Jun-wei,TAN Wei-jun,et al.Intelligent Recognition of Coal Gangue Based on Residual Network and Near Infrared Spectroscopy Technology[J].Journal of Instrumental Analysis,2024,43(04):607-613.
王亚栋,贾俊伟,谭韦君等.基于深度残差网络和近红外光谱的煤矸石智能识别[J].分析测试学报,2024,43(04):607-613. DOI: 10.12452/j.fxcsxb.23112615.
WANG Ya-dong,JIA Jun-wei,TAN Wei-jun,et al.Intelligent Recognition of Coal Gangue Based on Residual Network and Near Infrared Spectroscopy Technology[J].Journal of Instrumental Analysis,2024,43(04):607-613. DOI: 10.12452/j.fxcsxb.23112615.
该文开发了一种融合近红外光谱技术与一维残差深度网络(1D-ResNet)的煤炭及矸石快速分类方法。为保证实验样本的多样性,从河南、河北、山东3省份的多个煤矿中采集了430个煤炭与矸石样本,并基于欧氏距离对异常样本予以剔除,以获得高质量的建模数据集。在此基础上,为准确捕捉煤炭和矸石与其光谱特征之间的复杂映射关系,构建了基于1D-ResNet的分类模型,可在有效解决梯度消失问题的同时深度挖掘煤炭与矸石的光谱特性,获得高精度的分析结果。五折交叉验证结果显示,该模型的平均准确率达96.26%,显著优于支持向量机和随机森林等传统机器学习算法。在训练集和测试集上,该模型的损失函数变化趋势表现出较高的一致性,说明模型具备良好的泛化能力。测试发现,模型处理每一百个样本的推理时间仅为16.230 ms,进一步突显了其在煤炭与矸石在线分选领域的优势和潜在应用价值。
This study innovatively developed a rapid classification method for coal and gangue,integrating near infrared spectroscopy technology with a one-dimensional residual network(1D-ResNet). To ensure the diversity of experimental samples,430 samples of coal and gangue were collected from multiple coal mines in provinces such as Henan,Hebei,and Shandong. Abnormal samples were eliminated based on Euclidean distance to obtain a high-quality dataset for modeling. Building on this,a 1D-ResNet-based classification model was constructed to accurately capture the complex mapping relationships between coal,gangue,and their spectral characteristics. This approach effectively solved the problem of gradient vanishing and deeply mined the spectral features of coal and gangue,resulting in highly accurate analysis. After five-fold cross-validation,the model achieved an average accuracy of 96.26%,significantly outperforming traditional machine learning algorithms such as support vector machines and random forests. The model demonstrated high consistency in the trend of loss function changes across both the training and test datasets,indicating good generalization ability. Tests revealed that the model processes every hundred samples in just 16.230 milliseconds,further highlighting its advantages and potential application value in the online sorting field for coal and gangue.
煤矸石识别近红外光谱深度学习残差网络
coal gangue identificationnear infrared spectroscopydeep learningresidual network
Wang L,Wang X,Li B. Measurement,2023,219:113244.
Wang L,Wang X,Li B,Wei D. Int. J. Coal Prep. Util.,2023,43(6):1119-1137.
Zhang Y C,Wang J S,Yu Z W,Zhao S,Bei G X. Measurement,2022,198:111415.
Yan P C,Sun Q S,Yin N N,Hua L L,Shang S H,Zhang C Y. Measurement,2022,188:110530.
Cao X G,Li Y,Wang P,Wu X D. J. Mine Atuom.(曹现刚,李莹,王鹏,吴旭东. 工矿自动化),2020,46(1):38-43.
Yin J,Zhu H,Zhu J,Zeng Q,Li L,Yang C. Physicochem. Probl. Miner. Process,2022,58(1):24-36.
Yang Y,Li D,Guo Y,Wang S,Zhao D,Chen W,Zhang H. Meas. Sci. Technol.,2023,35(1):015404.
Huo X S,Chen P,Dai J W,Wang H P,Liu D,Li J Y,Xu Y P,Chu X L. J. Instrum. Anal. (霍学松,陈瀑,戴嘉伟,王海朋,刘丹,李敬岩,许育鹏,褚小立. 分析测试学报),2022,41(9):1301-1313.
Yang E,Wang S B,Xuan T. J. Mine Atuom.(杨恩,王世博,宣统. 工矿自动化),2022,48(7):22-31.
Yu X H,Guo W,Wu N,Lei M. Infrared Phys. Technol.,2020,105:103180.
Zou L,Qiao J H,Yu X H,Chen X,Lei M. IEEE Trans. Artif. Intell.,2023,32:1-13.
Song L,Liu S J,Mao Y C,Li T Z. J. Northeast. Univ.:Nat. Sci. (宋亮,刘善军,毛亚纯,李天子. 东北大学学报:自然科学版),2017,38(10):1473-1476.
Xiao D,Li H,Sun X. ACS Omega,2020,5(40):25772-25783.
Li L J,Fan S X,Wang X W,Li R,Wen X,Wang L Y,Li B. J. Spectrosc. Spectral Anal. (李廉洁,樊书祥,王学文,李瑞,文小,王璐瑶,李博. 光谱学与光谱分析),2022,42(4):1250-1256.
Shahamiri S R. IEEE Trans. Neural Syst. Rehabil. Eng.,2021,29:852-861.
Lin D,Lin J Z,Zhao L,Wang Z J,Chen Z K. IEEE Trans. Geosci. Electron.,2022,60:1-13.
Yang C G,Yin J Q,Wu L Q,Zeng Q Y,Zhang L W. ACS Omega,2022,8(1):48-55.
Liu Y,Wang L,Li H F,Chen X. Inf. Fusion,2022,86/87:1-16.
Zou L,Xu S F,Zhu W M,Huang X,Lei Z H,He K. Sensors,2023,23(16):7296.
Ma Z Y,Chang D L,Xie J Y,Ding Y F,Wen S G,Li X X,Si Z W,Guo J. IEEE Trans. Veh. Technol.,2019,68(4):3224-3233.
0
浏览量
27
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构