逆变器高频谐波扰动电子信号自动检测方法
Automatic Detection Method for High-frequency Harmonic Disturbance Electronic Signals in Inverters
席晓林 1张新一 1周金涛 1黄浩1
作者信息
- 1. 国网冀北电力有限公司智能配电网中心,河北秦皇岛 066000
- 折叠
摘要
研究逆变器高频谐波扰动电子信号自动检测方法.构建逆变器的电子信号采样函数,获取电子信号采样结果.利用改进后的阈值函数对电子信号进行小波去噪处理,通过混合字典对去噪后的信号进行稀疏分解,确定电子信号高频谐波扰动特征.利用BFGS校正拟牛顿算法优化BP神经网络的梯度和参数,将电子信号高频谐波扰动特征输入优化BP神经网络中,得到逆变器高频谐波扰动电子信号自动检测结果.实验结果表明,所提方法能精准获取逆变器电子信号,对高频谐波扰动电子信号的检测误差小,效果好.
Abstract
Research on automatic detection method for high-frequency harmonic disturbance electron-ic signals in inverters is conducted.The electronic signal sampling function of the inverter is constructed and the electronic signal sampling results are obtained.Using the improved threshold function to perform wavelet denoising on electronic signals,sparse decomposition is performed on the denoised signals through a hybrid dictionary to determine the high-frequency harmonic disturbance characteristics of electronic sig-nals.Using the BFGS correction quasi Newton algorithm to optimize the gradient and parameters of the BP neural network,the high-frequency harmonic disturbance characteristics of the electronic signal are input into the optimized BP neural network,and the automatic detection results of the high-frequency harmonic disturbance electronic signal of the inverter are obtained.The experimental results show that the proposed method can accurately obtain the electronic signals of the inverter,and the detection error of high-fre-quency harmonic disturbance electronic signals is low,with good results.
关键词
逆变器/高频谐波扰动/电子信号检测/改进阈值函数/混合字典/优化BP神经网络Key words
inverter/high frequency harmonic disturbance/electronic signal detection/improve the threshold function/hybrid dictionary/optimize BP neural network引用本文复制引用
基金项目
国网冀北电力有限公司科技投入项目(B7014422001M)
出版年
2024
国家科技期刊平台
NSTL
万方数据