TF-C Pretrain FD-B

- Paper: Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistency - Paper link: - Github repo: https://github.com/mims-harvard/TFC-pretraining - Project website: <br> <strong>FD-A</strong> and <strong>FD-B</strong> are subsets taken from the <strong>FD</strong> dataset, which is gathered from an electromechanical drive system that monitors the condition of rolling bearings and detect damages in them. There are four subsets of data collected under various conditions, whose parameters include rotational speed, load torque, and radial force. Each rolling bearing can be undamaged, inner damaged, and outer damaged, which leads to three classes in total. We denote the subsets corresponding to condition A and condition B as Faulty Detection Condition A (<strong>FD-A</strong>) and Faulty Detection Condition B (<strong>FD-B</strong>) , respectively. Each original recording has a single channel with sampling frequency of 64k Hz and lasts 4 seconds. To deal with the long duration, we followe the procedure described by Eldele et al., that is, we use sliding window length of 5,120 observations and a shifting length of either 1,024 or 4,096 to make the final number of samples relatively balanced between classes.

论文：基于时频一致性的时间序列自监督对比预训练
论文链接：无
GitHub 仓库：https://github.com/mims-harvard/TFC-pretraining
项目网站：无

FD-A 与 FD-B 均为故障检测（Faulty Detection，简称 FD）数据集的子集，该数据集采集自一套用于监测滚动轴承（rolling bearings）状态并检测其损伤的机电驱动系统。该数据集包含四组于不同采集工况下获取的子数据集，其采集参数涵盖转速、负载转矩与径向力。每一颗滚动轴承可分为完好、内圈损伤及外圈损伤三种状态，总计三类样本。我们将对应工况 A 与工况 B 的子数据集分别命名为故障检测工况 A（Faulty Detection Condition A，简称 FD-A）与故障检测工况 B（Faulty Detection Condition B，简称 FD-B）。每一段原始录制数据为单通道格式，采样频率为 64kHz，时长为 4 秒。为处理过长的采样时长，我们遵循 Eldele 等人提出的处理流程：采用长度为 5120 个观测值的滑动窗口，并以 1024 或 4096 作为滑动步长，最终使各类别的样本数量相对均衡。