Scholarlink

<p><b>−</b>Distributed drive electric mining trucks (DDEMTs) operate under continuous steep slopes with large load variations, where accurate driving condition recognition and state parameter estimation are essential for achieving efficient and reliable real-time torque control. This paper presents an integrated framework that combines Hidden Markov Model (HMM)–based driving condition recognition, Recursive Least Squares with Forgetting Factor (RLS-FF)–based truck mass estimation, Extended Kalman Filter (EKF)–based road slope estimation, and a Condition-Adaptive Energy Optimal Strategy (CA-EOS) for torque distribution. The HMM identifies three representative mining truck driving conditions—unloaded uphill, flat road, and loaded downhill—using the average motor current and activation ratio as adaptive time window features. The hierarchically coupled estimates of truck mass and road slope are incorporated as known inputs into the torque distribution, which constrains the required longitudinal force and driving torque. The optimization objective is adaptively formulated according to the recognized driving condition, ensuring that the distributed wheel torques satisfy minimizing total motor energy consumption under varying driving conditions. Both Matlab/Simulink simulations and Hardware-in-the-Loop (HIL) experiments verify that the proposed framework achieves accurate driving condition recognition and state parameter estimation, reducing total motor energy consumption by 11.68% over a transport cycle without compromising truck stability or real-time feasibility.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimization of Torque Distribution for Distributed Drive Electric Mining Trucks on Variable Road Surfaces

  • 期刊论文

  • 发表日期：

    2026年2月2日

查看全文
  • Yilin Wang,
  • Weiwei Yang,
  • Nong Zhang,
  • Haiping Du

摘要

−Distributed drive electric mining trucks (DDEMTs) operate under continuous steep slopes with large load variations, where accurate driving condition recognition and state parameter estimation are essential for achieving efficient and reliable real-time torque control. This paper presents an integrated framework that combines Hidden Markov Model (HMM)–based driving condition recognition, Recursive Least Squares with Forgetting Factor (RLS-FF)–based truck mass estimation, Extended Kalman Filter (EKF)–based road slope estimation, and a Condition-Adaptive Energy Optimal Strategy (CA-EOS) for torque distribution. The HMM identifies three representative mining truck driving conditions—unloaded uphill, flat road, and loaded downhill—using the average motor current and activation ratio as adaptive time window features. The hierarchically coupled estimates of truck mass and road slope are incorporated as known inputs into the torque distribution, which constrains the required longitudinal force and driving torque. The optimization objective is adaptively formulated according to the recognized driving condition, ensuring that the distributed wheel torques satisfy minimizing total motor energy consumption under varying driving conditions. Both Matlab/Simulink simulations and Hardware-in-the-Loop (HIL) experiments verify that the proposed framework achieves accurate driving condition recognition and state parameter estimation, reducing total motor energy consumption by 11.68% over a transport cycle without compromising truck stability or real-time feasibility.