基于改进Fast-ICA的电力谐波检测算法

摘要

In order to increase the detection speed and accuracy of Fast-ICA algorithm on harmonic,the original Fast-ICA blind source separation algorithm has been improved,so that iterations are reduced,and the convergence rate of the algorithm is raised,as well as the mathematical justification and simulation study are conducted.The testing results show that the algorithm average speed is increased by 37％ for the simple harmonic mixed signal and approximate the harmonic wave in the power grid signal detection with sampling frequency 7 500 Hz,920 sampling points.%为了提高Fast-ICA算法对电力谐波的检测速度与精度,在原有Fast-ICA盲源分离算法的检测原理上,对用于数据处理的二阶牛顿迭代法进行改进,使其满足三阶收敛,并给出收敛性证明.通过计算机仿真,在采样频率为7 500 Hz、采样点为920的实验条件下,利用该改进算法和原有算法分别对两组含谐波混合信号进行检测.仿真结果表明,改进算法的测量速度平均提高了37％,精确度也得到了较大提高.