【摘要】 针对辙机数量众多、工作环境恶劣、诸多因素导致设备故障频率高的问题，为实现对铁路转辙机故障的准确诊断，通过分析研究转辙机运行时产生的振动信号，提出了一种基于动态权重粒子群（DPSO）算法优化自适应神经模糊网络（ANFIS）的转辙机故障诊断方法。首先，利用集合经验模态分解（EEMD）算法将工况振动信号分解为若干本征模态函数（IMFs）并进行筛选；然后，使用改进时域多尺度散布熵（ITMDE）算法对IMFs提取特征熵值，进而输入经优化的ANFIS模型中学习实现故障诊断；最后，与多种诊断模型算法及学习算法进行对比分析。实验结果表明：本文方法可有效诊断转辙机故障，对转辙机故障智能诊断与日后相关研究具有一定参考意义。更多还原

【Abstract】 The number of railway signal switch machines is large, the working environment is bad, and many factors lead to the high frequency of equipment failure. In order to realize the accurate fault diagnosis of railway switch machine, a fault diagnosis method of switch machine based on dynamic weight particle swarm optimization and adaptive neural fuzzy network is proposed by analyzing the vibration signals generated during the switch machine operation. Firstly, the set empirical mode decomposition algorithm was used to decompose the vibration signals into several intrinsic mode functions and screen them. Then, the improved time-domain multi-scale spread entropy algorithm was used to extract the eigenentropy of IMFs, and then input the optimized ANFIS model to learn the fault diagnosis. Finally, it is compared with a variety of diagnostic model algorithms and learning algorithms. The experimental results show that the proposed method can effectively diagnose the fault of the switch machine, and has certain reference significance for the intelligent fault diagnosis of the switch machine and related research in the future.更多还原