Energy-Saving Strategy for Hybrid Tractors Based on Working Condition Prediction and DDPG-Fuzzy Control

Strategic Insights -

Researchers from Henan University of Science and Technology have proposed an innovative energy-saving strategy for hybrid tractors, leveraging the integration of working condition prediction and Deep Deterministic Policy Gradient and Fuzzy control (DDPG-Fuzzy) to optimize energy consumption. The study establishes a hybrid tractor system dynamics model, predicting working conditions and constructing a two-layer control architecture to optimize diesel benchmark torque and fuzzy control compensation. The results demonstrate significant reductions in fuel consumption and state of charge while maintaining performance under complex working conditions.

Key Takeaways:

  • The researchers proposed an energy-saving strategy for hybrid tractors based on working condition prediction and DDPG-Fuzzy control, demonstrating a 9.7% reduction in equivalent fuel consumption and a 4.4% reduction in state of charge under plowing conditions.
  • The working condition prediction model achieved mean prediction accuracies of 97% for plowing velocity and 96.8% for plowing resistance, setting the foundation for optimal energy consumption.
  • The study emphasized the importance of coordinating diesel and motor operations within high-efficiency regions to enhance fuel economy under complex working conditions.
  • A two-layer control architecture was designed, consisting of an upper DDPG algorithm layer and a lower fuzzy control compensation layer, effectively adjusting power allocation and torque correction.
  • The research team consisted of Shilong Fan, Xianghai Yan, Shuaishuai Ge, Junjiang Zhang, and Mengnan Liu from Henan University of Science and Technology.

Statistics:

  • The proposed energy-saving strategy demonstrated a 9.7% reduction in equivalent fuel consumption under plowing conditions.
  • The state of charge was reduced by 4.4% while maintaining it within a reasonable range.
  • The working condition prediction model achieved mean prediction accuracies of 97% for plowing velocity and 96.8% for plowing resistance.

Sources:

  • Research on Energy Saving for Hybrid Tractor Based on Working Condition Prediction and DDPG-Fuzzy Control. World Electric Vehicle Journal, 2025,16(9):490. (World Electric Vehicle Journal - http://www.mdpi.com/journal/wevj)
  • MDPI AG (publisher)
  • DOI: 10.3390/wevj16090490 (free version available at https://doi-org.sdpl.idm.oclc.org/10.3390/wevj16090490)