• Spectroscopy and Spectral Analysis
  • Vol. 41, Issue 7, 2288 (2021)
Peng-cheng YAN1、*, Song-hang SHANG2、2; *;, Chao-yin ZHANG2、2;, and Xiao-fei ZHANG2、2;
Author Affiliations
  • 11. State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mine, Anhui University of Science and Technology, Huainan 232001, China
  • 22. College of Electrical and Information Engineering, Anhui University of Science and Technology, Huainan 232001, China
  • show less
    DOI: 10.3964/j.issn.1000-0593(2021)07-2288-06 Cite this Article
    Peng-cheng YAN, Song-hang SHANG, Chao-yin ZHANG, Xiao-fei ZHANG. Classification of Coal Mine Water Sources by Improved BP Neural Network Algorithm[J]. Spectroscopy and Spectral Analysis, 2021, 41(7): 2288 Copy Citation Text show less
    Flow chart of mine water source prediction and classification model
    Fig. 1. Flow chart of mine water source prediction and classification model
    Original spectrum
    Fig. 2. Original spectrum
    Original spectrogram and preprocessed image(a): SG; (b): MSC; (c): Original preprocessed
    Fig. 3. Original spectrogram and preprocessed image
    (a): SG; (b): MSC; (c): Original preprocessed
    Cumulative contribution rate of principal components
    Fig. 4. Cumulative contribution rate of principal components
    Score distribution
    Fig. 5. Score distribution
    Comparison between the real value and the predicted value of each prediction model under nine different pretreatment methods
    Fig. 6. Comparison between the real value and the predicted value of each prediction model under nine different pretreatment methods
    Absolute error diagram of predicted value and true value of each model
    Fig. 7. Absolute error diagram of predicted value and true value of each model
    Fitness curves
    Fig. 8. Fitness curves
    隐节点数56789101112131415
    训练误差1.9141.3384.0665.2301.0360.0390.0380.0390.0370.0390.039
    Table 1. Training error of neuron number in different hidden layers
    预处理方式评价指标BP
    模型
    PSO-BP
    模型
    IPSO-BP
    模型
    R20.984 50.999 80.999 9
    SG预处理MRE0.073 90.001 70.000 1
    RMSE0.072 50.000 80.000 1
    R20.996 20.983 50.999 0
    MSC预处理MRE0.028 80.019 70.002 3
    RMSE0.023 50.066 90.003 9
    R20.995 50.998 80.998 8
    OriginalMRE0.044 10.009 10.004 2
    RMSE0.024 90.005 40.004 9
    Table 2. Comparison of evaluation indexes of each prediction model under different pretreatment methods
    Peng-cheng YAN, Song-hang SHANG, Chao-yin ZHANG, Xiao-fei ZHANG. Classification of Coal Mine Water Sources by Improved BP Neural Network Algorithm[J]. Spectroscopy and Spectral Analysis, 2021, 41(7): 2288
    Download Citation