基于决策树模型的用电负荷异常模式智能识别方法
By constructing a method for identifying abnormal power load patterns using a decision tree model, this method solves the problem of joint analysis of multiple load gap scenarios, achieves quantitative identification of reversals, jumps, and deviations, and provides explainable causes of anomalies and rapid verification support.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUNAN ZHONGQINGNENG TECH CO LTD
- Filing Date
- 2026-04-21
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies lack joint analysis of multiple gap scenarios when identifying abnormal power load patterns, making it difficult to identify reverse sequence problems, accurately reflect differences in user adjustability, and lack quantification of the amplitude of jumps in adjacent scenarios and the overall degree of deviation. Their classification capabilities are insufficient and it is difficult to explain the causes of anomalies.
A decision tree-based approach is adopted. By constructing a comparable daily set, a baseline load and low-level boundary are formed. The maximum adjustable depth is calculated, a reference control boundary is generated, multi-dimensional boundary features are constructed, and a monotonically adjustable boundary decision tree is used for cluster localization and anomaly screening to output anomaly pattern categories.
It achieves robust identification of multiple gap scenarios, accurately reflects the user's adjustable capabilities, can quantify the degree of reversal, jump and deviation, provide explainable reasons for anomalies, and support rapid verification and correction.
Smart Images
Figure CN122072682B_ABST