Tsallis小波包奇异熵与功率谱分析在电力谐波检测的应用

Application of Wavelet Packet Singularity Entropy and PSD in Power Harmonics Detection

针对Shannon小波能量熵在电力谐波检测过程中的局限性,提出一种基于Tsallis小波包奇异熵与功率谱(PSD)分析结合的电力谐波检测新方法。从Shannon熵和Mallat算法理论入手,论证Shannon小波能量熵在电力谐波复杂度表征及频率辨析中存在的不足。在分析Tsallis熵与Shannon熵的区别与联系的基础上,将Tsallis非广延熵理论和小波包算法结合,构造具有非广延性的Tsallis小波包奇异熵。根据Tsallis小波包奇异熵对电力谐波复杂度的表征结果,对谐波信号进行分段PSD分析。理论分析及仿真结果证明:该方法在正确表征电力谐波复杂度的同时,能对电力谐波的具体频率和功率进行具体量化。

Aiming at the limitations of Shannon wavelet energy entropy （WEE）,a novel approach to detect power harmonics,based on Tsallis wavelet packet singularity entropy （WPSE） and power spectral density （PSD） analysis,is proposed. By means of Shannon entropy theory and Mallat algorithm,the shortcomings of Shannon WEE applied to characterize complexity and frequency features of power harmonics are discussed in detail. After analyzing the relations and differences between Tsallis entropy and Shannon entropy,Nonextensive Tsallis WPSE algorithm is provided by combining Tsallis entropy with wavelet packet transform（WPT）. According to the complexity of power harmonics calculated by Tsallis WPSE,the harmonics are analyzed in different time stages by PSD. The theoretical analysis and simulation results indicate that the proposed approach is valid to characterize the complexity of harmonics. In addition,the frequency and power of harmonics are measured precisely.