To improve the accuracy and sensitivity of power information communication data traffic anomaly detection,the power information communication data traffic anomaly detection method based on value derivative gated recurrent unit(GRU)is proposed.The wavelet transform method is used to reduce the noise of power information communication data traffic.Utilizing the improved clustering algorithm,the noise reduction power information communication data traffic is clustered.The clustered power information communication data traffic is input into.the value derivative GRU model to realize the power information communication traffic anomaly detection.The experimental results show that the method has better noise reduction effect of power information communication data traffic and can effectively improve the accuracy and sensitivity of power information communication data traffic anomaly detection.The method can be used in network communication and Internet of Things,which is of great significance to the development of information and communication technology.
关键词
电力信息通信/数据流量/值导数门控循环单元/异常检测/小波变换/改进聚类算法
Key words
Power information communication/Data traffic/Value derivative gated recurrent unit(GRU)/Anomaly detection/Wavelet transform/Improved clustering algorithm
维普
万方数据