基于改进QPSO形态滤波器的电能质量暂态扰动检测  

Power Quality Transient Disturbance Detection Based on Improved QPSO Morphological Filter

作　　者：盛四清[1] 王佳琦[1] 

机构地区：[1]华北电力大学电气与电子工程学院,河北保定071003

出　　处：《中国电力》2017年第6期95-100,107,共7页Electric Power

摘　　要：提出一种基于改进量子粒子群算法(QPSO)的形态滤波器,用于电能质量暂态扰动检测。首先对量子粒子群算法进行改进,利用混沌序列对粒子位置初始化以提高算法全局寻优能力,通过引入变异算子进一步避免算法过早收敛。然后将改进后的算法应用于形态学滤波器的结构元素的自适应优化中,结合结构元素的特点,寻找出属性最佳的结构元素从而提高滤波能力。结合仿真实验,研究了含有复杂多变噪声的环境下,发生电压骤升、电压骤降和电压中断暂态干扰现象时所构造改进滤波器的滤波性能。通过实验对比,证明所提方法具有快速、准确的特点,较传统的滤波方法有了很大的改善,提高了电能质量扰动检测的可靠性。A morphological filter based on improved quantum particle swarm optimization （QPSO） algorithm is proposed for power quality transient disturbance detection. Firstly, the quantum particle swarm optimization algorithm is improved. Particle position is initialized by chaos sequence to improve global optimization ability, and the mutation operator is introduced to avoid premature convergence. Then, the improved algorithm is applied to adaptive optimization of structural elements of morphological filter, which is combined with characteristics of structural elements to find out the best structural elements of the attributes to improve the filtering ability. Through simulation experiment of containing complex and changeful environment noise, voltage swell, voltage sag, voltage interruption phenomenon of transient interference constructed and performance of the improved filter is researched. Compared with the experimental results, it is proved that proposed method is fast and accurate. Compared with traditional filtering method, it also improves reliability of power quality disturbance detection.

关 键 词：数学形态学 量子粒子群算法 电能质量扰动 结构元素 滤波 

分 类 号：TM711[电气工程—电力系统及自动化] TM930