Long-short-time domain torque optimal prediction and allocation method for electric logistics vehicles with electro-hydraulic composite steering system

Abstract

The electro-hydraulic composite steering system can reduce steering energy consumption through cooperative control of electro-hydraulic mechanisms, an inevitable trend for global commercial vehicles and green logistics. However, in cooperative control, the electro-hydraulic coupling characteristics not only lead to an increase in system energy consumption but also cause fluctuations in system speed during electro-hydraulic switching. In response to the above issues, this paper proposes a long-short-time domain steering mode selection and torque optimal allocation strategy that integrates long-time domain steering mode selection and short-time domain torque allocation. In the long-term domain, with steering energy consumption as the optimization indicator, the optimal steering mode is selected through a steering mode selection model based on the CNN-LSTM network to reduce steering energy consumption. In the short time domain, the Holt Winter exponential smoothing and support vector regression methods are combined for torque prediction, and the steering energy consumption and electro-hydraulic switching smoothness indicators are comprehensively considered. The electro-hydraulic torque distribution ratio is dynamically optimized in the rolling time domain to reduce the fluctuation of hydraulic pump speed during electro-hydraulic switching. The simulation and experimental results show that the proposed method can improve the switching smoothness of the electro-hydraulic composite steering system and reduce the system’s overall energy consumption by 54.1%.