论文部分内容阅读
由于间谐波具有在频域广泛分布、幅值相对于基波和谐波较小、周期难以确定等特点,因此间谐波的检测难度很高,要求能够尽量限制非同步采样对检测精度的影响。现代谱估计算法由于对同步采样没有要求,频率分辨率高等特点被广泛地用于电力系统谐波与间谐波检测中,由于其只能准确的估计信号的频率,不能很好的估计信号的幅值和相位,在实际应用中受到很大限制,而支持向量机如果要准确的估计信号的频率、幅值和相位,计算量很大。基于以上特点,将支持向量机(SVM)与总体最小二乘旋转不变(TLS-ESPRIT)算法结合起来,先用TLS-ESPRIT算法准确估计信号频率分量,再用支持向量机方法估计信号的幅值和相位,通过三个算例分别讨论了该方法在相邻信号、多信号等情况下的检测情况,证明该方法切实可行,能准确检测出间谐波的各种参数,并且比单独使用支持向量机计算量小。
Interharmonics are widely distributed in the frequency domain, the amplitude is small relative to the fundamental and harmonic, the period is difficult to determine and so on, so the detection of interharmonics is very difficult, and it is required to limit the non-synchronous sampling to the detection accuracy influences. Modern spectrum estimation algorithm is not required for synchronous sampling, high frequency resolution characteristics are widely used in power system harmonics and inter-harmonic detection, because it can only accurately estimate the frequency of the signal can not be a good estimate of the signal The magnitude and phase are very limited in practical application. However, if the SVM accurately estimates the frequency, amplitude and phase of the signal, the calculation is very large. Based on the above features, the support vector machine (SVM) is combined with the total least squares rotation invariant (TLS-ESPRIT) algorithm. The signal frequency components are first estimated by TLS-ESPRIT algorithm and the signal amplitude is estimated by the support vector machine Value and phase of the method, the detection of the method under adjacent signal, multiple signals and so on are discussed through three examples respectively. The method is proved to be practical and feasible, and various parameters of interharmonics can be accurately detected, Support vector machine calculation is small.