基于ZD6转辙机电流数据的道岔转辙器健康状态划分

Health Status Classification of Switch Based on ZD6 Switch Machine Current Data

为了能够准确评估道岔转辙器的健康状态,进而实现道岔转辙器“状态修”,以ZD6转辙机驱动的道岔转辙器为研究对象,采集ZD6转辙机在额定牵引力的100%-120%区间内不同负载下的工作电流数据构建数据集,在此数据集上采用CNN-SVM网络进行健康状态划分。首先利用CNN对电流数据进行特征提取,然后将CNN提取的特征输入到SVM中进行训练和分类,从而实现对ZD6转辙机牵引的道岔转辙器健康状态的划分。实验结果表明,该方法的分类准确率达到93.3%,为铁路系统的安全运维提供了有力的数据支撑。

To accurately assess the health state of turnout rutters,and enable“condition-based maintenance”,this study focuses on ZD6 switch machine-driven turnout rutters.A dataset was constructed by collecting operating current of ZD6 rutters under different loads within the range of 100-120%of the rated traction force,A CNN-SVM network was then applied to classify the health status based on this dataset.Firstly,CNN was utilized to extract features from the current data,and then the extracted features were inputted into SVM for training and classification,thereby achieving the classification of the health status of the turnout rutters driven by ZD6 rutting machine.Experimental results demonstrate that the classification accuracy of this method reaches 93.3%,verifing the effectiveness of the model and providing strong data support for the safe operation and maintenance of railway system.