Study on Prediction Model of Soil Cadmium Content Moisture Content Correction Based on GWO-SVR

Ying Chen; Can Zhang; Chunyan Xiao; Xueliang Zhao; Yanxin Shi; Hui Yang; Zhengying Liu; Shaohua Li

doi:10.3788/AOS202040.1030002

Acta Optica Sinica, Volume. 40, Issue 10, 1030002(2020)

Study on Prediction Model of Soil Cadmium Content Moisture Content Correction Based on GWO-SVR

Ying Chen^1、*, Can Zhang¹, Chunyan Xiao², Xueliang Zhao^1,3, Yanxin Shi³, Hui Yang¹, Zhengying Liu¹, and Shaohua Li⁴

¹Hebei Province Key Laboratory of Test/Measurement Technology and Instrument, School of Electrical Engineering, Yanshan University, Qinhuangdao, Hebei 0 66004, China

²Institute of Resources & Environment, Henan Polytechnic University, Jiaozuo, Henan 454000, China;

³Center for Hydrogeology and Environmental Geology, China Geological Survey, Baoding, Hebei 0 71051, China

⁴Hebei Sailhero Environmental Protection High-tech Co., Ltd., Shijiazhuang, Hebei 0 50035, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(10)

Fig. 1. XRF under different experimental conditions. (a) Under different cadmium contents; (b) under different moisture contents

Download full size

Fig. 2. Comparison of characteristic peaks before and after Gaussian function fitting

Download full size

Fig. 3. Flow chart of GWO-SVR algorithm

Download full size

Fig. 4. Optimization process of sample box model and data comparison. (a) Iterative optimization process; (b) comparison between prediction data and test data

Download full size

Fig. 5. Optimization process of tablet model and data comparison. (a) Iterative optimization process; (b) comparison between prediction data and test data

Download full size

Fig. 6. Comparison between prediction data and real data. (a) Data of sample box model; (b) data of tablet model

Download full size

Table 1. Data comparison of sample box model
View table
Table 1. Data comparison of sample box model
Model R² MSE MAE
ENR 0.9425 0.0551 0.2157
Lasso 0.9424 0.0551 0.2158
RR 0.9233 0.0827 0.2642
SVR 0.9672 0.0354 0.1746
GWO-SVR 0.9753 0.0185 0.1502

Table 2. Data comparison of tablet model
View table
Table 2. Data comparison of tablet model
Model R² MSE MAE
ENR 0.9094 0.0706 0.2656
Lasso 0.9095 0.0711 0.2536
RR 0.8774 0.1137 0.3058
SVR 0.9611 0.0359 0.1640
GWO-SVR 0.9837 0.0093 0.0979

Table 3. Data comparison of sample box SVR optimized model
View table
Table 3. Data comparison of sample box SVR optimized model
Model R² MSE MNI
PSO-SVR 0.9653 0.0371 14
GA-SVR 0.9559 0.0522 17
SA-SVR 0.9684 0.0327 7
GWO-SVR 0.9753 0.0185 4

Table 4. Data comparison of tablet SVR optimized model
View table
Table 4. Data comparison of tablet SVR optimized model
Model R² MSE MNI
PSO-SVR 0.9675 0.0207 7
GA-SVR 0.9689 0.0195 9
SA-SVR 0.9759 0.0107 4
GWO-SVR 0.9837 0.0093 4

Tools

Get Citation

Copy Citation Text

Ying Chen, Can Zhang, Chunyan Xiao, Xueliang Zhao, Yanxin Shi, Hui Yang, Zhengying Liu, Shaohua Li. Study on Prediction Model of Soil Cadmium Content Moisture Content Correction Based on GWO-SVR[J]. Acta Optica Sinica, 2020, 40(10): 1030002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites