Data processing method for current transformer test
By employing least squares fitting, symlet wavelet denoising, Chirp Z-transform, and harmonic flux compensation models, the problem of incomplete error reflection in current transformer testing was solved, achieving high-precision performance evaluation of current transformers.
Patent Information
- Application Number
- CN202511331558.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-18
- Publication Date
- 2025-10-21
- Estimated Expiration
- 2045-09-18
AI Technical Summary
Existing current transformer testing methods cannot accurately reflect the comprehensive error of the transformer under harmonic or transient conditions. Traditional methods cannot effectively simulate complex operating conditions, and signal processing suffers from spectral leakage, noise interference, and core nonlinearity errors, resulting in insufficient test accuracy.
The least squares method was used to fit and subtract the exponentially decaying aperiodic components. Denoising was performed by combining Symlet wavelet and Stan unbiased risk estimation. High-resolution spectrum analysis was performed using Chirp Z-transform, and a harmonic flux compensation model was established for error compensation.
It improves the accuracy and reliability of current transformer ratio error and angle error testing, overcomes spectrum leakage and noise interference, accurately compensates for core nonlinearity error, and enhances testing precision.