Discharge fertilizer rate prediction across fertilizer type for a bivariate fertilization system based on transfer learning

Authors

  • Jiqin Zhang School of Mechanical and Automotive Engineering, Anhui Polytechnic University, Wuhu 241000, Anhui, China
  • Qibin Zhuang School of Mechanical and Automotive Engineering, Anhui Polytechnic University, Wuhu 241000, Anhui, China
  • Lin Xi School of Mechanical and Automotive Engineering, Anhui Polytechnic University, Wuhu 241000, Anhui, China
  • Gang Liu Key Laboratory of Smart Agriculture System Integration Research of the Ministry of Education, China Agricultural University, Beijing 100083, China
  • Zhao Zhang Key Laboratory of Smart Agriculture System Integration Research of the Ministry of Education, China Agricultural University, Beijing 100083, China
  • Changyuan Zhai Intelligent Equipment Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China
  • Shuo Yang Intelligent Equipment Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China

Keywords:

discharge fertilizer rate (FDR) prediction, bivariate fertilizer applicator (BFA), decision tree regression (DTR), transfer learning (TL), instance-based transfer learning

Abstract

Accurate prediction of discharge fertilizer rate (DFR) is essential for achieving precise control in bivariate fertilizer applicators (BFA). However, most existing prediction models calibrated for specific fertilizer types ignore variations in material properties, limiting their generalizability and necessitating time-consuming recalibration for different fertilizers. Transfer learning (TL) offers a promising solution by leveraging knowledge from a source domain (DS) to improve performance in a related target domain (DT). This study investigated the cross-fertilizer DFR prediction in a BFA system using two transfer scenarios: from a compound fertilizer (Stanley) to another (Sakefu), and from Stanley to urea. Decision tree regression (DTR) serves as the base model. Five experimental configurations were compared: direct source domain prediction (DTR- DS); partial target domain prediction with 20% data (DTR-; joint training (DTR-Recalibration); instance-based TL using transfer adaptive boosting for regression (TrAdaBoost.R2) (DTR-TrAdaBoost.R2); and full target domain prediction (DTR-DT). The results demonstrate that: 1) DTR-Recalibration is most effective for Stanley to Sakefu, improving R2 by 2.4% and reducing nRMSE by 12.5%, while DTR-TrAdaBoost.R2 performs best for Stanley to urea, increasing R2 by 14.3% and reducing nRMSE by 21.5%; 2) differences in material properties significantly affect TL performance; the ranking of the importance of material characteristics is as follows: the combined difference in equivalent diameter and angle of repose > bulk density > sphericity ratio; and 3) high-weight instances in TL are predominantly concentrated at medium operational intensity levels, and the stability of the discharge ratio () critically influences transfer effectiveness. These findings provide practical and theoretical insights for cross-fertilizer model transfer, enabling efficient adaptation of DFR prediction models with minimal target data and reducing calibration effort in precision agriculture applications.      

Key words: discharge fertilizer rate (DFR) prediction; bivariate fertilizer applicator (BFA); decision tree regression (DTR); transfer learning (TL); instance-based transfer learning

DOI: 10.25165/j.ijabe.20261904.10387

Citation: Zhang J Q, Zhuang Q B, Xi L, Liu G, Zhang Z, Zhai C Y, et al. Discharge fertilizer rate prediction across fertilizertype for a bivariate fertilization system based on transfer learning. Int J Agric & Biol Eng, 2026; 19(4): 51–66.

References

[1] Wang X W, Jiang R, Zhou Z Y, Song C C, Luo X W, Bao R F, et al. Discharge rate consistency of each channel for UAV-based pneumatic granular fertilizer spreader. Int J Agric & Biol Eng, 2023; 16(4): 20–28.

[2] Toselli M, Baldi E, Ferro F, Rossi S, Cillis D. Smart farming tool for monitoring nutrients in soil and plants for precise fertilization. Horticulturae. 2023; 9(9): 1–14. doi: 10.3390/horticulturae9091011

[3] Pawase P P, Nalawade S M, Bhanage G B, Walunj A A, Kadam P B, Durgude A G, et al. Variable rate fertilizer application technology for nutrient management: A review. Int J Agric & Biol Eng, 2023; 16(4): 11–19.

[4] Beneduzzi H M, de Souza E G, Moreira W K O, Sobjak R, Bazzi CL, Rodrigues M. Fertilizer recommendation methods for precision agriculture-a systematic literature study. Engenharia Agricola. 2022; 42(1). doi: 10.1590/1809-4430-eng.agric.v42n1e20210185/2022

[5] Alameen A A, Al-Gaadi K A, Tola E K. Development and performance evaluation of a control system for variable rate granular fertilizer application. Computers and Electronics in Agriculture, 2019; 160: 31–39.

[6] Zhang J Q, Liu G. Effects of control sequence optimisation on the performance of bivariate fertiliser applicator. Computers and Electronics in Agriculture, 2022; 192(17): 106594.

[7] Zhang J Q, Zhuang Q B, Liu G. Analysis of the multi-objective control sequence optimization problem in bivariate fertilizer applicators. Symmetry, 2025; 17(6): 926.

[8] Su N, Xu T S, Song L T, Wang R J, Wei Y Y. Variable rate fertilization system with adjustable active feed-roll length. Int J Agric & Biol Eng, 2015; 8(4): 19–26.

[9] Zhang J, Liu G, Luo C, Hu H, Huang J. MOEA/D-DE based bivariate control sequence optimization of a variable- rate fertilizer applicator. Computers and Electronics in Agriculture, 2019; 167: 105063.

[10] Shi Y Y, Xin Y P, Wang X C, Zheng E L, Shen C, Zhang Z. Research on fertilizer application strategy for rice-wheat dual-variable precision fertilizer applicator based on MLP. Transactions of the CSAE, 2025; 41(10): 51–60. (in Chinese) doi: 10.11975/j.issn.1002-6819.202408134

[11] Hossen M I, Awrangjeb M, Pan S, Mamun A A. Transfer learning in agriculture: A review. Artificial Intelligence Review, 2025; 58(4). doi: 10.1007/s10462-024-11081-x

[12] Narmadha R P, Sengottaiyan N, Kavitha R J. Deep transfer learning based rice plant disease detection model. Intelligent Automation and Soft Computing, 2022; 31(2): 1257–1271.

[13] Panigrahi S, Nanda A, Swarnkar T. A survey on transfer learning. Smart Innovation, Systems and Technologies, 2021; 194: 781–789.

[14] Wang Y C, Zhang S Q, Dai B S, Yang S S, Song H C. Fine-grained weed recognition using Swin Transformer and two-stage transfer learning. Frontiers in Plant Science, 2023; 14: 1–13.

[15] Krishnamoorthy N, Narasimha Prasad L V, Pavan Kumar C S, Subedi B, Abraha H B, Sathishkumar V E. Rice leaf diseases prediction using deep neural networks with transfer learning. Environmental Research, 2021; 198: 111275.

[16] Yin X, Li W H, Li Z, Yi L L. Recognition of grape leaf diseases using MobileNetV3 and deep transfer learning. Int J Agric & Biol Eng, 2022; 15(3): 184–194. Int J Agric & Biol Eng, 2022; 15(3): 184–194. doi: 10.25165/j.ijabe.20221503.7062

[17] Ma Z, Wang Y, Zhang T S, Wang H G, Jia Y J, Gao R, et al. Maize leaf disease identification using deep transfer convolutional neural networks. Int J Agric & Biol Eng, 2022; 15(5): 187–195.

[18] Ma Y C, Zhang Z, Yang H L, Yang Z W. An adaptive adversarial domain adaptation approach for corn yield prediction. Computers and Electronics in Agriculture, 2021; 187: 106314.

[19] Lim H-J, Kim Y-J, Kim S-M, Kim S-J. Convolutional Neural Network (CNN)-based transfer learning framework for cherry tomato production. Int J Agric & Biol Eng, 2025; 18(5): 90–101.

[20] Chen Y, Li L, Whiting M, Chen F, Sun Z C, Song K S, et al. Convolutional neural network model for soil moisture prediction and its transferability analysis based on laboratory Vis-NIR spectral data. International Journal of Applied Earth Observation and Geoinformation, 2021; 104: 102550.

[21] Yang J, Li J T, Hu J, Yang W J, Zhang X L, Xu J F, et al. An interpretable deep learning approach for calibration transfer among multiple near-infrared instruments. Computers and Electronics in Agriculture, 2022; 192: 106584.

[22] Pan S J, Yang Q. A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineeringngineering, 2009; 22: 1345–1359.

[23] Wu S T, Zhu Y S, Zhang Q. A new robust visual tracking algorithm based on transfer adaptive boosting. Mathematical Methods in the Applied Sciences, 2012; 35(17). doi: 10.1002/mma.2643

[24] Huang Y R, Chen W Q, Tan W, Deng Y J, Yang C H, Zhu X G, et al. Transfer learning for enhancing the generality of leaf spectroscopic models in estimating crop foliar nutrients across growth stages. International Journal of Applied Earth Observation and Geoinformation, 2025; 139. doi: 10.1016/j.jag.2025.104481

[25] Bu H R, Yu S Y, Dong W C, Wang Y Q, Zhang L X, Xia Y Q. Calibration and testing of discrete element simulation parameters for urea particles. Processes, 2022; 10(3). doi: 10.3390/pr10030511

[26] Miller T, Mikiciuk G, Kisiel A, Mikiciuk M, Paliwoda D, Sas-Paszt L, et al. Machine learning approaches for forecasting the best microbial strains to alleviate drought impact in agriculture. Agriculture (Switzerland), 2023; 13(8): 1–16.

[27] Torres-Sanchez R, Navarro-Hellin H, Guillamon-Frutos A, San-Segundo R, Ruiz-Abellón M C, Domingo-Miguel R. A decision support system for irrigation management: Analysis and implementation of different learning techniques. Water (Switzerland). 2020;12(2): 548. doi:10.3390/w12020548

[28] Biddapp.C B, Srikanth V. Crop yield prediction on agriculture using machine learning. International Journal of Research Publication and Reviews, 2024; 5(3): 165–167.

[29] Chen L, Cui Y P, Wang M, Niu Y C, Xu A G, Liu K Y, et al. A study of relationships between cultivated land fertility and corn head smut using a decision tree model. Journal of Agricultural Resources and Environment, 2021; 38: 928–36.

[30] Paudel D, de Wit A, Boogaard H, Marcos D, Osinga S, Athanasiadis I N. Interpretability of deep learning models for crop yield forecasting. Computers and Electronics in Agriculture, 2023; 206: 107663.

[31] Tang D J, Yang X H, Wang X S. Improving the transferability of the crash prediction model using the TrAdaBoost. R2 algorithm. Accident Analysis and Prevention, 2020; 141: 105551.

[32] Dai W Y, Yang Q, Xue G R, Yu Y. Boosting for transfer learning. In: Proceedings of the 24th international conference on Machine learning. Corvalis, Oregon, USA; 2007; pp.193–200. doi: 10.1145/1273496.1273521

[33] Skobalski J, Sagan V, Alifu H, Al Akkad O, Lopes F A, Grignola F. Bridging the gap between crop breeding and GeoAI: Soybean yield prediction from multispectral UAV images with transfer learning. ISPRS Journal of Photogrammetry and Remote Sensing, 2024; 210: 260–281.

Downloads

Published

2026-09-03

How to Cite

(1)
Zhang, J.; Zhuang, Q.; Xi, L.; Liu, G.; Zhang, Z.; Zhai, C.; Yang, S. Discharge Fertilizer Rate Prediction across Fertilizer Type for a Bivariate Fertilization System Based on Transfer Learning. Int J Agric & Biol Eng 2026, 19, 51-66.

Issue

Vol. 19 No. 4 (2026): IJABE

Section

Applied Science, Engineering and Technology

License

Copyright (c) 2026 International Journal of Agricultural and Biological Engineering

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

IJABE is an international peer reviewed, open access journal, adopting Creative Commons Copyright Notices as follows.

Authors who publish with this journal agree to the following terms:

  1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
  2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
  3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).