HCL-MTSAD: Hierarchical Contrastive Consistency Learning for Accurate Detection of Industrial Multivariate Time Series Anomalies

from arxiv, This paper is a manuscript that is still in the process of revision, including Table 1, Figure 2, problem definition in section III.B and method description proposed in section IV. In addition, the submitter has not been authorized by the first author and other co-authors to post the paper to arXiv

Multivariate Time Series (MTS) anomaly detection focuses on pinpointing samples that diverge from standard operational patterns, which is crucial for ensuring the safety and security of industrial applications. The primary challenge in this domain is to develop representations capable of discerning anomalies effectively. The prevalent methods for anomaly detection in the literature are predominantly reconstruction-based and predictive in nature. However, they typically concentrate on a single-dimensional instance level, thereby not fully harnessing the complex associations inherent in industrial MTS. To address this issue, we propose a novel self-supervised hierarchical contrastive consistency learning method for detecting anomalies in MTS, named HCL-MTSAD. It innovatively leverages data consistency at multiple levels inherent in industrial MTS, systematically capturing consistent associations across four latent levels-measurement, sample, channel, and process. By developing a multi-layer contrastive loss, HCL-MTSAD can extensively mine data consistency and spatio-temporal association, resulting in more informative representations. Subsequently, an anomaly discrimination module, grounded in self-supervised hierarchical contrastive learning, is designed to detect timestamp-level anomalies by calculating multi-scale data consistency. Extensive experiments conducted on six diverse MTS datasets retrieved from real cyber-physical systems and server machines, in comparison with 20 baselines, indicate that HCL-MTSAD's anomaly detection capability outperforms the state-of-the-art benchmark models by an average of 1.8\% in terms of F1 score.

翻译：多变量时间序列（MTS）异常检测聚焦于识别偏离标准运行模式的样本，这对确保工业应用的安全性至关重要。该领域的主要挑战在于开发能够有效辨别异常的表征。现有文献中主流的异常检测方法本质上是基于重建和预测的。然而，它们通常专注于单一维度的实例层面，未能充分利用工业多变量时间序列中固有的复杂关联。为解决此问题，我们提出了一种新颖的自监督分层对比一致性学习方法（HCL-MTSAD）用于检测多变量时间序列异常。该方法创新性地利用工业多变量时间序列中多个层面上的数据一致性，系统性地捕获了四个潜在层面（测量、样本、通道和过程）的一致关联。通过开发多层对比损失函数，HCL-MTSAD能够深度挖掘数据一致性与时空关联，生成更具信息量的表征。进而，基于自监督分层对比学习设计异常判别模块，通过计算多尺度数据一致性来检测时间戳级别的异常。在来自真实信息物理系统和服务器的六个多样化多变量时间序列数据集上开展的大量实验，与20个基线模型的对比表明，以F1分数衡量，HCL-MTSAD的异常检测能力平均比最先进的基准模型高出1.8%。