Predicting Sugarcane Yield via the Use of an Improved Least Squares Support Vector Machine and Water Cycle Optimization Model

Yifang Zhou, Mingzhang Pan, Wei Guan, Changcheng Fu and Tiecheng Su ()
Additional contact information
Yifang Zhou: State Key Laboratory for the Protection and Utilization of Subtropical Agricultural Biological Resources, College of Mechanical Engineering, Guangxi University, Nanning 530004, China
Mingzhang Pan: State Key Laboratory for the Protection and Utilization of Subtropical Agricultural Biological Resources, College of Mechanical Engineering, Guangxi University, Nanning 530004, China
Wei Guan: State Key Laboratory for the Protection and Utilization of Subtropical Agricultural Biological Resources, College of Mechanical Engineering, Guangxi University, Nanning 530004, China
Changcheng Fu: State Key Laboratory for the Protection and Utilization of Subtropical Agricultural Biological Resources, College of Mechanical Engineering, Guangxi University, Nanning 530004, China
Tiecheng Su: State Key Laboratory for the Protection and Utilization of Subtropical Agricultural Biological Resources, College of Mechanical Engineering, Guangxi University, Nanning 530004, China

Agriculture, 2023, vol. 13, issue 11, 1-23

Abstract: As a raw material for sugar, ethanol, and energy, sugarcane plays an important role in China’s strategic material reserves, economic development, and energy production. To guarantee the sustainable growth of the sugarcane industry and boost sustainable energy reserves, it is imperative to forecast the yield in the primary sugarcane production regions. However, due to environmental differences caused by regional differences and changeable climate, the accuracy of traditional models is generally low. In this study, we counted the environmental information and yield of the main sugarcane-producing areas in the past 15 years, adopted the LSSVM algorithm to construct the environmental information and sugarcane yield model, and combined it with WCA to optimize the parameters of LSSVM. To verify the validity of the proposed model, WCA-LSSVM is applied to two instances based on temporal differences and geographical differences and compared with other models. The results show that the accuracy of the WCA-LSSVM model is much better than that of other yield prediction models. The RMSE of the two instances are 5.385 ton/ha and 5.032 ton/ha, respectively, accounting for 7.65% and 6.92% of the average yield. And the other evaluation indicators MAE, R 2 , MAPE, and SMAPE are also ahead of the other models to varying degrees. We also conducted a sensitivity analysis of environmental variables at different growth stages of sugarcane and found that in addition to the main influencing factors (temperature and precipitation), soil humidity at different depths had a significant impact on crop yield. In conclusion, this study presents a highly precise model for predicting sugarcane yield, a useful tool for planning sugarcane production, enhancing yield, and advancing the field of agricultural production prediction.

Keywords: crop production; agricultural production; artificial intelligence; machine learning; parameter optimization; sensitivity analysis (search for similar items in EconPapers)
JEL-codes: Q1 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2077-0472/13/11/2115/pdf (application/pdf)
https://www.mdpi.com/2077-0472/13/11/2115/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jagris:v:13:y:2023:i:11:p:2115-:d:1276068

Access Statistics for this article

Agriculture is currently edited by Ms. Leda Xuan

More articles in Agriculture from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().