基于Hermite多项式展开的变压器局部放电信号类型识别方法
By combining Hermite polynomial expansion and support vector machine, the accuracy problem of multi-source partial discharge signal type identification is solved, and high-precision discharge type classification is achieved, which is suitable for the identification of transformer partial discharge signals.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGZHOU POWER SUPPLY BUREAU GUANGDONG POWER GRID CO LTD
- Filing Date
- 2022-08-08
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for identifying partial discharge signal types struggle to accurately separate and identify signals from multiple sources concurrently, especially in complex environments such as substations. Existing algorithms cannot effectively handle the mixing of multiple PD signals, resulting in poor separation performance.
A Hermite polynomial expansion-based method is adopted to approximate the partial discharge signal through Hermite polynomial series. The orthogonality and recursiveness of Hermite polynomials are used to extract feature vectors, which are then combined with support vector machines for classification and recognition. The order and scaling factor of the Hermite polynomials are optimized to improve the signal classification accuracy.
It improves the accuracy of partial discharge signal classification and enables high-precision discharge type identification in noisy environments. Experimental results show that the identification rate reaches over 97%, effectively solving the problem of separating multi-source partial discharge signals.
Smart Images
Figure CN115345200B_ABST