【摘要】 针对中低速磁悬浮列车具有较强非线性且运行环境未知干扰多等问题，提出一种基于机理与数据驱动的中低速磁悬浮列车建模方法。基于实际磁悬浮线路现场获取的数据利用变分模态分解(VMD)进行去噪处理，以获取平稳的时间序列；对列车进行机理分析建立列车的动态模型，运用带遗忘因子的最小二乘法辨识线性模型参数，同时利用TCN-BiLSTM估计未建模动态部分；设计磁悬浮列车混合智能补偿模型，为评估其性能和泛化能力，将该模型应用于中低速磁悬浮列车的不同工况(牵引、巡航、制动)进行仿真。试验结果表明，该模型不仅能精准描述列车运行过程中的复杂非线性关系及动态变化，同时在各种工况条件下均展现出了精准的拟合能力，具有较好的理论意义和应用价值。更多还原

【Abstract】 Aiming at the problems of strong nonlinearity and numerous unknown disturbances in the operating environment of medium and low-speed maglev trains, a modeling method based on mechanism and data-driven approaches was proposed for medium and low-speed maglev trains. Variational mode decomposition(VMD)was used to denoise the data obtained from the actual maglev line to obtain a stable time series. A dynamic model of the train was developed by analyzing the mechanism of the train. The forgetting factor recursive least square(FFRLS)was used to parameterize the linear model, while the TCN-BiLSTM was used to estimate the unmodeled dynamic parts. A hybrid intelligent compensation model was designed for maglev trains. In order to evaluate its performance and generalization ability, the model was applied to different operating conditions(traction, cruising, braking) of medium and low speed maglev trains for simulation. The experimental results show that the model can not only accurately describe the complex nonlinear relationships and dynamic changes in the train operation process, but also shows accurate fitting ability under various operating conditions, which has good theoretical significance and application potential.更多还原