基于ICA和MI的谐波源识别研究

Study on harmonic source identification based on ICA and MI

把电力系统谐波源识别问题抽象为盲源分离问题。运用独立分量分析和互信息论,提出了一种电网多谐波源识别的新方法。独立分量分析法用于重构谐波电流,互信息论用于谐波源的定位。为证明所述方法的正确性,运用IEEE39节点建立仿真来识别多谐波源。结论证明在网络参数未知的条件下,所提出的方法能准确识别多个谐波源的位置。

The harmonic source identification problem of power system is abstracted to blind source separation in this paper. Using independent component analysis and information theory, a new method for power grid multi harmonic sources identification is proposd. Independent component analysis method is used to reconstruct the harmonic current while mutual information theory is used to locate the harmonic source. In order to prove the exactness of the method, the IEEE39 node is used to make a simulation to identify multi harmonic sources. Conclusion proves that under the unknown network parameters conditions, the proposed method can accurately identify the location of the multi harmon- ic sources.