In response to the prediction error caused by frequency fluctuations in railway signal transmission,this article proposes a deep learning based anomaly interference prediction method.This method comprehensively obtains node interference data,extracts key features using data processing techniques,and reflects signal anomalies.Based on deep learning algorithms,construct an interference signal anomaly detection framework to automatically learn and recognize abnormal patterns.Through real-time data input,the framework can quickly output prediction results and accurately predict abnormal interference.Experimental results have shown that the average prediction error distance of this method is 50.5 kHz,which is more than 85 kHz lower than the control group.It achieves accurate prediction of abnormal interference in railway signal transmission nodes and provides strong guarantees for the safe and stable operation of railways.
关键词
深度学习模型/铁路信号/传输节点/异常干扰预测/时域偏度
Key words
deep learning model/railway signal/transmission node/abnormal interference prediction/time domain skewness
维普
万方数据