工程车辆自动变速挡位决策的遗传径向基神经网络方法

为了能通过4个挡位的控制使液力变矩器在高效区工作,提高工程车辆自动变速传动系统的效率,利用径向基函数(RBF)神经网络较强的输入输出映射功能提出了一种基于径向基函数神经网络的工程车辆自动变速控制方法。以ZL50E装载机传动试验台换挡控制试验的数据为样本,采用遗传算法对RBF神经网络进行训练,并进行了验证性的仿真试验。仿真结果表明:该方法能够根据车辆运行状态确定最佳挡位,从而及时、准确地满足工程车辆自动换挡的要求。

RBF Neural Network with Genetic Algorithm of Shift Decision for Automatic Transmission of Construction Vehicle

Keeping the torque converter working in the high efficiency range controlled by the four shifting gears to improve the efficiency of the automatic transmission system of the construction vehicles, an approach of its control based on the RBF neural network was proposed to take advantage of the strong mapping function between the inputs and the outputs of such a network. The RBF neural network was trained by the genetic algorithm, taking the experimental data of the shifting control over ZL50E wheel loader at test bench as samples, and the verifying simulational tests were performed. The simulation results show that the optimal shifting gear can be decided by the proposed approach, and the requirement of the construction vehicle to the automatic shifting can be satisfied in time and accurately.