摘要
利用模糊隶属度函数的概念,提出了一种新的电能质量检测去噪算法,该算法针对电能质量信号的每个点设计了1×5模板和基于该模板的5个掩模,利用最大隶属度原则选取一个掩模,对该点进行去噪,对每个点应用此算法直至完毕。实验仿真结果表明,该算法在去噪和突变点信息保留方面的效果远远优于传统的中值滤波和均值滤波,拥有良好的去噪能力和保留突变点信息的能力。本算法新颖、便捷,有广阔的发展前景。
As familiar de-noising methods, median filter and average fiher both have many disadvantages. By using fuzzy theory and membership function, a new denoising method of power quality detection is proposed. For each point of power quality signal, this method designs 1 × 5 template and 5 cover templates. By using the most membership principle, it selects a cover template and carries on the denoising to the point. Use this method to each point. The simulation experiment result indicates that, for the effects of denoising and keeping the brake points＇information, this method is much better than median filter and average filter. It has better ability of denoising and keeping the break points＇information. This method is novel and convenient with wide developmental foreground.
出处
《继电器》
CSCD
北大核心
2006年第6期44-48,共5页
Relay
基金
国家自然科学基金(No.60402004)
关键词
电能质量检测
隶属度函数
去噪
中值滤波
均值滤波
power quality detection
membership function
de-noising
median filter
average fiher
