异常点检测方法、装置及存储介质、电子设备
By performing sliding window processing and adaptive parameter calculation on the power load sequence, and dynamically adjusting the detection threshold, the problem of high false detection rate and low efficiency in anomaly detection in power load sequence data is solved, and more efficient anomaly identification is achieved.
CN115238735BActive Publication Date: 2026-07-17CHINA GRIDCOM +1
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA GRIDCOM
- Filing Date
- 2022-06-24
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing technologies for detecting anomalies in electricity load sequence data suffer from high false detection rates and low efficiency.
Method used
A sliding window-based and adaptive time series outlier detection method is adopted. By performing sliding window processing on the electricity load series, the skewness value, baseline and parameters of the subsequence are calculated, and the detection threshold is dynamically adjusted to identify data outliers.
Benefits of technology
It improves the accuracy and efficiency of anomaly detection and reduces the false detection rate, especially significantly improving detection efficiency under large-scale power load sequence data.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN115238735B_ABST
Abstract
本发明公开了一种异常点检测方法、装置及存储介质、电子设备，该方法包括：获取用电负荷序列，对用电负荷序列进行滑窗处理，得到多个子序列；计算每一子序列的偏度值，并根据偏度值确定对应子序列的基线和参数；根据基线和参数对对应子序列中的数据进行检测，得到每一子序列中的数据异常点；根据所有子序列中的数据异常点，得到用电负荷序列的异常点检测结果。本发明的异常点检测方法，是基于滑窗和自适应的时间序列异常点检测方法，具有异常检测更合理，检测效率更高优点。
Need to check novelty before this filing date? Find Prior Art