【摘要】 在局部放电在线监测和实时信号处理中,干扰较大时单纯依靠传统的硬件滤波较难准确提取有用信号,而在软件抗干扰中,传统多抽头自适应滤波器的参数设置也较难,有时会导致收敛很缓慢甚至不稳定。为此,提出了一种将二进小波变换与自适应滤波结合的局放信号检测算法,先对信号作简单的单抽头自适应LMS算法滤波,然后对滤波结果进行基于āTrous算法的二进小波变换来提取有用信号。通过对仿真白噪、窄带周期干扰及现场实测数据处理表明,该算法有很好的去噪性能,稳定性和实时性优于传统的自适应滤波,满足大部分工程要求,为工程实际应用提供了有效的去噪手段。更多还原

【Abstract】 Anti-interference is always an important and difficult problem in online monitoring of electric power facilities.There is many kinds of noises in engineering locale where various electric facilities is working,and at many times the amplitude of the noise may be greater than real partial discharge signals.In this time,it is difficult to eliminate noises and get the partial discharge signals by traditional hardware devices,and it is not only difficult in setting parameters of traditional self-adaptive filters but also the result may not be constringency.So it would prevent insulated fault of electric power facilities from monitoring in time.In this paper,a new way of anti-interference of partial discharge based on the binary wavelet transform and adaptive LMS algorithm is proposed,dealing signals which conclude real partial discharge signals and noises with self-adaptive filter based on LMS algorithm and has only one offset at first,then doing binary wavelet transform of the result of the first step above.Emulation and study of engineering locale signals shows that the new way has better performance and stability compared to usual self-adaptive filters,and the time which need to deal with signals of one power cycle in recent sample frequency is within ten seconds,it can satisfy most engineering need.So it can be a good way to the engineering use in detecting partial discharge signal.更多还原