论文部分内容阅读
针对电力系统谐波污染问题,提出了基于特征值分解和快速独立分量分析(FastICA)的谐波/间谐波检测算法。该方法在不需要任何先验知识的情况下,将单道电力系统混合信号通过时间延迟构造出多道观测信号,对其自相关函数进行特征值分解确定原谐波/间谐波信号中源信号频率成分的个数,确定观测信号矩阵的阶数,再利用FastICA算法对谐波/间谐波信号中各个频率成分进行分离提取,借助频谱分析得到各个成分的频率估计。在此基础上,借助最小二乘法对谐波/间谐波信号进行幅值和相位估计。通过仿真实验与其他经典算法比较,充分说明了所提出算法的可行性、准确性和有效性。
Aiming at the problem of harmonic pollution in power system, harmonic / interharmonic detection algorithm based on eigenvalue decomposition and fast independent component analysis (FastICA) is proposed. The proposed method not only needs any prior knowledge, but also constructs a multichannel observer signal through the time delay of the mixed signal of single-channel power system. The autocorrelation function is decomposed to determine the source of the original harmonic / interharmonic signals Determine the order of the matrix of the observed signal, and then use the FastICA algorithm to separate and extract each frequency component in the harmonic / interharmonic signals, and obtain the frequency estimation of each component by means of spectrum analysis. Based on this, the amplitude and phase of harmonic / interharmonic signals are estimated by means of least square method. Compared with other classical algorithms through simulation experiments, the feasibility, accuracy and validity of the proposed algorithm are fully illustrated.