基于大数据的高速列车气动载荷作用下迭代学习主动控制研究

列车在高速运行的过程中与另一列车相交会时,将产生剧烈的瞬态气动载荷冲击而引起车体瞬间横向振动加剧,导致列车横向平稳性恶化。为了改善列车运行平稳性,采用大数据方法及迭代学习控制思想,提出基于高速列车运行大数据的迭代学习主动控制算法,并进行多体动力学与控制算法的联合仿真,进一步研究控制算法对会车气动载荷幅值变化和会车时间变化的鲁棒性。结果表明：大数据迭代学习主动控制经过5次迭代后对会车气动载荷激扰下的车体横向振动峰值降低52.67%,且控制算法对会车工况变化有较好的鲁棒性。

Research on Big Data Based Iterative Learning Active Control on High Speed Train under Aerodynamic Loads

Strong transient aerodynamic loads will produced when a train passing another in high speed. The aerodynamic loads cause transient vehicle lateral vibration, which causes the deterioration of lateral stability. To improve the train running stability, big data method and iterative learning control were used and iterative learning active control algorithm based on high-speed train running big data was proposed. The co-simulations based on multibody dynamics and control algorithm were performed and even researched robustness of control algorithm when the magnitude of aerodynamic loads or intersection time varied. The results show that the big data based iterative learning active control through five iterations can make vehicle lateral vibration peak reduce 52.67% under aerodynamic loads, and control algorithm has strong robustness when intersection conditions change.