Online adaptive shift control method for heavy-duty electric commercial vehicles considering road slope and shift frequency

To address the problems of cyclic shifting, poor working condition adaptability and low economy in the automated manual transmission shifting for heavy-duty electric commercial vehicles, this study adopted the “offline optimization + online application” method and proposed an online adaptive shifting control method for heavy-duty electric commercial vehicles (HD-ECV) considering road slope and shifting frequency. This study takes a 6X4 HD-ECV as the research object. Firstly, under typical commercial vehicle driving cycles, an offline-optimized gear shift control strategy was developed based on the dynamic programming algorithm, targeting minimization of whole-vehicle energy consumption through the incorporation of gear shifting penalty terms. Secondly, based on the optimal gear obtained offline and combined with the vehicle operating conditions, an optimal gear selection rule table is generated. Then, according to the optimal gear selection rule table, the gear selection model is trained based on the convolutional neural network-bidirectional long short-term memory model with vehicle velocity, acceleration, road slope, and wheel-end torque as input and gear as output. Finally, the trained neural network model is loaded into the vehicle controller to control the vehicle's gear selection online. The results show that the proposed control method can effectively reduce the number of gears shifting and the energy consumption when applied online. Furthermore, the control methodology's effectiveness was validated through hardware-in-the-loop testing. This study provides new ideas and methods for the design of shift strategies for HD-ECV.