基于长度可变结构元素数学形态学的谐波检测算法研究

传统数学形态学滤波算法中,当结构元素过长时,会对滤波算法的延时性以及硬件成本等方面造成不良影响。针对此种局限性,提出了一种结构元素变长度的数学形态学滤波算法,即通过对腐蚀和膨胀运算加入比例系数进行改进,以改进后所形成的腐蚀和膨胀新运算法则构建均值组合滤波器,使改进后的数学形态滤波器能够在保证精度的同时,大幅度降低滤波器对结构元素长度的需求。最后,将改进后的形态滤波器应用到以瞬时无功功率理论为基础的谐波检测算法中,仿真结果表明改进后的算法具有存储量需求低和快速响应的性能。

Research on the Harmonic Detection Algorithm Based on Mathematical Morphological with Length-variable Structure Element

In traditional mathematical morphological filtering algorithm, the over-length of structure element has an adverse effect on filtering algorithm latency, hardware cost and other aspects. Due to the limitation, a modified mathematical morphological filtering algorithm with length-variable structure element is presented in this paper, which is modified by adding proportionality coefficient to the erosion and dilation algorithm. The mean filter is established on the new algorithm which is based on modified erosion and dilation operation. As a result, the precision of the modified mathematical morphological filter is guaranteed, meanwhile the length of the modified mathematical morphological filter is reduced sharply. Finally, the modified mathematical morphological filter is applied to the harmonic detection algorithm based on instantaneous reactive power theory. Simulation indicates that the modified filter has the performance of the low need of storage and rapid response.