PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 23, 2026Transactions of the Canadian Society for Mechanical Engineering0 citations

Data-driven Traction Characteristics Modeling and Prediction of Agricultural Tractor Transmission Systems Using Physics-informed Neural Network Approach

View Full Paper
XYXianghai YanMSMengyao ShiHWHang Wang

Key Points

  • The central aim is to improve the prediction of agricultural tractor traction performance using a Physics-informed Neural Network (PINN) approach.
  • Developed a Physics-informed Neural Network framework integrating neural networks with traction dynamics knowledge.
  • Incorporated physics-based constraints into the loss function to ensure physical consistency during training.
  • Evaluated the model's performance against traditional empirical methods with experiments on tractor traction outputs.
  • Achieved an average coefficient of determination (R2) above 95% across four output indicators.
  • Demonstrated high accuracy in predicting traction power and fuel consumption rates.
  • Observed trends predicted by the model showed strong consistency with actual measured data.

Abstract

Agricultural tractor traction performance prediction is crucial for reducing fiel consumption and improving operation quality in modern agriculture. To address the limitations of traditional empirical modeling and regression methods in predicting tractor traction performance, where model accuracy and computational efficiency are often constrained by multidimensional nonlinear characteristics, this study introduces a Physics-informed Neural Network (PINN) framework. The proposed approach integrates the nonlinear approximation capability of neural networks with prior knowledge from traction dynamics, and incorporates physics-based constraint terms into the loss function to enhance physical consistency and generalization. Unlike conventional black-box data-driven models, the PINN framework simultaneously fits observed data and enforces consistency with physical laws during training, which prevents physically implausible predictions. Experimental results show that the PINN achieves high accuracy, particularly in predicting traction power and fuel consumption rate, with an average coefficient of determination (R2) above 95% across four output indicators. In addition, the predicted trends are highly consistent with measured data, demonstrating the potential of this method to provide strong technical support for the optimization and design of agricultural machinery power systems.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Yan et al. (2026) studied this question.

synapsesocial.com/papers/69e9b89b85696592c86ebb8bhttps://doi.org/10.1139/tcsme-2025-0213
Ask AI
Helpful
Bookmark
Share
View Full Paper