Research on Prediction Model of Soil Heavy Metal Zn Content Based on XRF-CNN
CHEN Ying1, YANG Hui1, XIAO Chun-yan2, ZHAO Xue-liang1, 3, LI Kang3, PANG Li-li3, SHI Yan-xin3, LIU Zheng-ying1, LI Shao-hua4
1. Hebei Province Key Laboratory of Test/Measurement Technology and Instrument, School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China
2. School of Resources and Environment, Henan University of Technology, Jiaozuo 454000, China
3. Center for Hydrogeology and Environmental Geology, China Geological Survey, Geological Environment Monitoring Engineering Technology Innovation Center of The Ministry of Natural Resources, Baoding 071051, China
4. Hebei Sailhero Environmental Protection High-tech Co., Ltd., Shijiazhuang 050035, China
Abstract:Combined with X-ray fluorescence spectroscopy, a prediction model based on deep convolutional neural network regression is proposed to predict the content of heavy metal element Zn in soil. Related pretreatment of the original soil, and soil compaction by powder compaction method, and the soil spectrum was obtained by X-Ray-fluorescence (XRF). Compared with traditional detection methods, the XRF method has the advantages of fast detection speed, high accuracy, simple operation, non-destructive sample properties, and simultaneous detection of multiple heavy metal elements. Therefore, XRF is combined with deep convolutional neural networks to achieve Precise prediction of heavy metal element Zn content in soil. In the experiment, box plots were used to eliminate abnormal data in the X-ray fluorescence spectrum. Entropy weight method and multiple scattering correction were used to correct the sample box data. The Savitzky-Golay smooth denoising method and linear background method are used to preprocess the spectral data, which can effectively solve the problems of noise and baseline drift caused by the external environment and human factors. The obtained one-dimensional spectral data vector is processed by constructing a spectral data matrix, this method converts 5 sets of parallel spectral data vectors at the same concentration and the same water content into a two-dimensional spectral information matrix and uses this matrix as the input of the deep convolutional neural network prediction model to meet the operational requirements of the convolutional layer. The learning ability of the deep convolutional neural network prediction model is improved, and the training difficulty of the model is reduced. The deep convolutional neural network prediction model is built with 3 layers of convolutional layers and activated using the RELU activation function. The maximum pooling method is used to reduce the dimensionality of the data and increase the Dropout layer to prevent overfitting. The ADAM optimizer is used to optimize the prediction model. The prediction model uses the mean relative error (MRE), loss function (LOSS), and mean absolute error (MAE) to determine the optimal learning rate of the model is and the optimal number of iterations is 3 000. The prediction model of the deep convolutional neural network is compared with the BP prediction model, the ELM prediction model, and the PLS prediction model, Analyze and compare the prediction model with the mean square error (MSE), root mean square error (RMSE), and fitting coefficient (R2), The results show that in predicting the content of heavy metals in soil, the prediction model based on deep convolutional neural network is superior to the three prediction models of BP, ELM, and PLS, which improves the prediction accuracy.
Key words:Soil heavy metals; X-ray fluorescence spectroscopy; Spectral information matrix; Deep convolutional neural network
陈 颖,杨 惠,肖春艳,赵学亮,李 康,庞丽丽,史彦新,刘峥莹,李少华. 基于XRF-CNN土壤重金属Zn元素含量预测模型研究[J]. 光谱学与光谱分析, 2021, 41(03): 880-885.
CHEN Ying, YANG Hui, XIAO Chun-yan, ZHAO Xue-liang, LI Kang, PANG Li-li, SHI Yan-xin, LIU Zheng-ying, LI Shao-hua. Research on Prediction Model of Soil Heavy Metal Zn Content Based on XRF-CNN. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(03): 880-885.
[1] Shard A G, Wright L, Minelli C. Biointerphases, 2018, 13(6): 061002.
[2] Tekin Z, Unutkan T, Erulaş F, et al. Food Chemistry, 2020,310: 125825.
[3] Krata A A, Wojciechowski M, Vassileva E, et al. Journal of Food Composition and Analysis, 2020, 86: 103381.
[4] Silva S H G, Weindorf D C, Pinto L C, et al. Geoderma, 2020, 362: 114136.
[5] Montanha G S, Rodrigues E S, Marques J P R, et al. Metallomics, 2020, 12: 183.
[6] Chen Y M, Zi Y Y, Cao H R, et al. Science China Technological Sciences, 2014, 57(3): 589.
[7] Ng W, Minasny B, McBratney A. Science of the Total Environment, 2020, 702: 134723.
[8] LI Ling-qiao, PAN Xi-peng, FENG Yan-chun, et al(李灵巧,潘细朋,冯艳春,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2019, 39(11): 3606.
[9] Akbarpour M R, Pouresmaeil A. Diamond and Related Materials, 2018, 88: 6.
[10] Perera E N C, Jayawardana D T, Jayasinghe P, et al. Modeling Earth Systems and Environment, 2019,5(4): 1635.
[11] Verbeke J M, Glenn A M, Keefer G J, et al. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2016, 825: 69.
[12] Romo-Cárdenas G, Avilés-Rodríguez G J, Sánchez-López J D, et al. Results in Physics, 2018, 11: 17.
[13] Stephens D, Smith A, Redfern T, et al. Applied Computing and Geosciences, 2020,5: 100016.