改进粒子群算法在电能质量信号去噪中的应用

摘要

For the defects and deficiencies in thresholding method of transient power quality disturbances signal, this paper is proposed a thresholding method based on improved PSO (Particle Swarm Optimal), used SURE unbiased estimate of adaptive optimal threshold selection method to select threshold, Introduce the correction parameter factor to improve inertia weight of PSO iteration function based on introduction of particle evolution speed factor and aggregation degree factor, which addresses the problem that PSO falls into local optimum in the late time, the threshold is more adaptive and the accuracy of optimal solution is improved, and the effect of de-noise is more obvious. Finally through simulation of de-noising, the result shows that the method proposed in this paper has better effect on de-noising than traditional Particle Swarm Optimal thresholding method, and has a better prospect.%针对暂态电能质量扰动信号阈值去噪方法的缺陷及不足，提出了基于改进粒子群的最优阈值法，采用基于SURE无偏估计的自适应最优阈值选择方法对阈值进行选取。在引入粒子群进化速度因子、聚集度因子的基础上加入参数修正因子对粒子群迭代函数的惯性权重进行改进，解决了粒子群算法在后期易陷入局部最优的问题，阈值自适应性及阈值求解精度得到提升。最后通过对两类常见电能质量扰动信号进行去噪仿真验证，结果表明该方法较传统粒子群阈值去噪方法的去噪效果更为明显，具有更好的应用前景。