Abstract
Anomaly detection methods are part of the systems where rare events may endanger an operation's profitability, safety, and environmental aspects. Although many state-of-the-art anomaly detection methods were developed to date, their deployment is limited to the operation conditions present during the model training. Online anomaly detection brings the capability to adapt to data drifts and change points that may not be represented during model development resulting in prolonged service life. This paper proposes an online anomaly detection algorithm for existing real-time infrastructures where low-latency detection is required and novel patterns in data occur unpredictably. The online inverse cumulative distribution-based approach is introduced to eliminate common problems of offline anomaly detectors, meanwhile providing dynamic process limits to normal operation. The benefit of the proposed method is the ease of use, fast computation, and deployability as shown in two case studies of real microgrid operation data.
Abstract (translated)
异常检测方法是系统中的一小部分,罕见事件可能会危及业务的盈利能力、安全性和环境方面。尽管至今已经开发出了许多先进的异常检测方法,但它们的部署局限于模型训练时存在的运行条件。在线异常检测带来了适应数据偏差和模型发展中可能未体现在数据中的改变点的能力,从而延长了服务寿命。本文提出了一种在线异常检测算法,适用于需要低延迟检测的现有实时基础设施,且数据中出现了不可预测的新模式。在线逆累积分布方法被引入,以消除 offline 异常检测常见的问题,同时为正常操作提供了动态过程限制。本文提出的方法的优点包括易于使用、快速计算和可部署性,实际 microgrid 运行数据的两个案例研究展示了这种方法的优势。
URL
https://arxiv.org/abs/2301.13527