高速列车纵向动力学建模与自适应RBFNN控制

Longitudinal Dynamics Modeling and Adaptive RBFNN Control for High-speed Trains

高速列车由多节车厢链接而成的结构特性导致其高速运行在变路况线路条件下难以有效地对其进行优化控制。针对上述问题,提出一种高速列车纵向动力学模型与径向基函数神经网络(RBFNN)控制策略。考虑列车车钩力和复杂线路条件,分析整列车前后的不同受力情况,建立列车纵向动力学模型。针对该模型无外加干扰时设计一种理想反馈控制律,引入RBFNN对理想控制输出进行拟合,在考虑干扰项影响的情况下,通过设计参数估计自适应律代替神经网络权值的调整,并对其进行Lyapunov稳定性证明。采用京石武高铁北京西—郑州东段的CRH380B型高速列车真实线路运行数据进行仿真模拟,并在相同条件下与反演滑模(BSSM)控制器的仿真结果进行对比。仿真结果表明所提控制器更能有效应对复杂路况变化和外界干扰,对高速列车具有更好的控制效果,改善其运行的平稳性及高效性。

Due to the structural characteristics of high-speed trains linked by multiple carriages,it is difficult to effectively control the high-speed trains during high speed operation under changing high-speed railway line conditions.In response to the above problems,this paper proposed a high-speed train longitudinal dynamics model and an adaptive radial basis function neural network(RBFNN)control strategy.Firstly,considering the train coupler force and complex line conditions,based on the analysis of the different forces in the front and rear of the whole train,the train longitudinal dynamics model was established.Secondly,an ideal feedback control law was designed for the model without external interferences,and RBFNN was introduced to fit the ideal control output.Then,the adaptive law of design parameter estimation was used to replace the adjustment of the weights of the neural network under the condition of considering the influence of the interference term,and the Lyapunov stability of the model was proved.Finally,the real line running data of the CRH380B high-speed train from the section between the Beijingxi Railway Station and Zhengzhoudong Railway Station was used for simulation,and the simulation results were compared with the backstepping sliding mode(BSSM)controller under the same conditions.The simulation results show that the proposed controller can more effectively deal with complex high-speed railway condition changes and external disturbances,has better control effect on high-speed trains,and improves the stability and efficiency of their operation.