电力系统中谐波检测是无功补偿、谐波滤除等电能质量分析与控制的前提。针对传统电力系统中谐波检测方法存在的问题,结合盖氏圆和TAM算法进行谐波检测。该算法以信号空间为模型,首先用盖氏圆检测出信号源数,然后结合信号子空间和噪声子空间的正交性以及子空间的旋转不变性,估计出信号的频率,最后由全最小二乘法(TLS)计算得到信号的幅值。仿真实验与实际数据分析结果表明:该方法速度快、精确度高,具有一定的应用前景。
In power system, harmonic detection is the basis of the analysis and control of power quality, e. g. , reactive power compensation and harmonic filtration. In order to overcome the shortcoming of the traditional methods, this paper proposes a harmonic detection method by combining Gerschgorin disk criterion and Toeplitz approximation method (TAM). This method is based on the models of signal spaces. Gerschgorin disk criterion is employed to detect the source number. Besides, both the orthogonality of signal sub-space and noise sub-space and the rotational invariance of subspaces are combined to estimate the signal frequency. And then, the signal amplitude is obtained by utilizing total least squares(TLS) algorithm. Finally, it is shown from simulation experiments and real data analysis that the proposed algorithm can attain higher speediness and better precision.