【摘要】 针对轨道电路设备应用场景多样且复杂电磁骚扰源影响轨道电路信号传输的问题，提出一种基于麻雀搜索算法（Sparrow Search Algorithm,SSA）的优化变分模态分解方法 （Variational Modal Decomposition,VMD），实现轨道电路信号分析处理。首先，通过基于平均包络熵适应度函数的麻雀搜索算法，实现VMD关键参数的优化选取；其次，采用优化参数的VMD方法，分离深度耦合的轨道电路信号和随机骚扰，实现强噪声背景下轨道电路信号的检测以及骚扰成分的提取和降维；最后，基于Matlab生成仿真混叠信号进行验证，对比提出的SSA-VMD轨道电路信号处理方法与现有信号自适应分解方法的处理效果。结果表明：SSA-VMD方法较现有方法在准确性上有较大优势，处理后信号的信噪比提升可达30 dB；同时，使用现场实测含噪数据验证也表明，SSA-VMD方法对于轨道电路信号的分析处理能达到预期的应用效果。更多还原

【Abstract】 Addressing the challenges posed by diverse application scenarios and the influence of complex electromagnetic disturbance sources on signal transmission of track circuit, a track circuit signal processing and analysis method based on Variational Modal Decomposition(VMD) is proposed based on the Sparrow Search Algorithm(SSA) for track circuit signal analysis and processing. This method first realizes the intelligent selection of key parameters of VMD utilizing the average envelope entropy as the fitness function. On this basis, the deeply coupled track circuit signal are separated from random disturbances, enabling the detection of track circuit signal under strong noise background, as well as the extraction and dimensionality reduction of disturbance components. Finally, simulation mixture signals are generated based on Matlab for validation.Compared with the processing effect of existing signal adaptive decomposition methods, SSA-VMD has higher accuracy, and the signal-to-noise ratio of the processed signal can be improved by up to 30 dB. Validation with field-measured noisy data also shows that the proposed method effectively meets the expected application requirements for track circuit signal analysis and processing.更多还原