关键词: 微电网, 谐波检测, 希尔伯特-黄变换（HHT）, 经验模态分解, 瞬时频率

Abstract: In order to solve the problem to accurately detect complex non-stationary signals, such as micro-grid harmonic and mutation, a method based on Hilbert-Huang Transform （HHT） is presented. The method is about harmonic detection and time-frequency analysis for micro-grid. It fits extreme value curve by adopting shape-preserving piecewise cubic Hermite interpolation, and implements Empirical Mode Decomposition （EMD） on harmonic signals, then a finite number of Intrinsic Mode Components （IMF） are obtained, and Hilbert transformation is conducted. The instantaneous frequency and amplitude of each IMF components are calculated finally. The time-frequency characteristics of micro-grid harmonic in non-stationary power signals are detected accurately. The simulation results show that the method can quickly and precisely obtain the harmonic signal frequency components, amplitude and voltage mutation time. Compared with FFT and traditional HHT methods, it has higher accuracy and obvious time distinguishing feature, which meets the engineering application requirements for micro-grid harmonic computer detection.

Key words: microgrid, harmonic detection, Hilbert-Huang Transform（HHT）, empirical mode decomposition, instantaneous frequency