Jake5/wmx_normal_data

该数据集旨在训练LSTM自编码器模型，以检测由WMX Ethercat主设备计算的DC diff统计数据的异常。数据集包含两列数据：Timestamp和DcDiffAvg，其中DcDiffAvg表示每个周期到下一个DC时钟的平均时间间隔，以纳秒为单位，并显示出特定的锯齿模式。通过该数据集，可以训练自编码器模型以检测主设备与子设备之间通信不稳定时的异常。

This dataset is used to train an LSTM autoencoder model to detect anomalies in DC diff statistics calculated by the WMX Ethercat master. The dataset consists of two columns: Timestamp and DcDiffAvg. After each cycle is completed, the average time interval to the next DC clock is calculated in nanoseconds, showing a peculiar sawtooth pattern. Using this dataset, the autoencoder model can be trained to detect unstable communication between the master and sub-devices.