郭中華 徐靜 魏菁 姬智
【摘 要】為了提高暫態(tài)電能質(zhì)量檢測的精度,就要求被檢信號去噪效果達到最佳;本文通過對比小波去噪和小波包去噪的優(yōu)劣性,以此選出效果較好的小波包去噪進行深層次篩選,包括對sym4小波包、sym8小波包、dB3小波包、dB5小波包去噪的效果比較,從而篩選出效果最好的sym5小波包去噪方法。該方法有效克服了小波去噪效果不佳而致使信號失真的缺點,應(yīng)用范圍也不僅僅局限在低頻范圍;除此之外,減少了去噪過程中信息損失,使信噪比和均方根誤差等性能指標(biāo)較之小波去噪有明顯優(yōu)化。
【關(guān)鍵詞】暫態(tài)電能質(zhì)量;電壓;小波去噪;小波包去噪;信噪比;均方根誤差
Wavelet packet denoising in the application of the transient power quality signal denoising
GUO Zhong-hua XU Jing WEI Jing JI Zhi
(1.School of Physics and Electronic Information Engineering, Yinchuan Ningxia 750021, China;
2.Ningxia University; Ningxia Key Laboratory of Intelligent Sensing for Desert Information, Yinchuan Ningxia 750021, China)
【Abstract】In order to improve the accuracy of transient power quality detection, requires that the effect of detected signal denoising is best; In this paper, by comparing the superiority-inferiority of wavelet denoising and wavelet packet denoising, wavelet packet denoising with better effect for deep filter, including the effect of sym4 wavelet packet, sym8 wavelet packet, dB3 wavelet packet and dB5 wavelet packet denoising,so as to screen out the best effect of sym5 wavelet packet denoising method. This method is effective to overcome the effect of wavelet denoising is not good and the shortcoming of the signal distortion, scope of application is not limited in the low frequency range. And reduced the loss of information, at the same time in the denoising of SNR and RSME performance index is obviously enhanced compared with the wavelet denoising.
【Key words】Transient power quality; Voltage; Wavelet denoising; Wavelet packet denoising; Signal-to-noise ratio; Root mean square error
0 引言
電能質(zhì)量問題分為穩(wěn)態(tài)電能質(zhì)量問題和暫態(tài)電能質(zhì)量問題兩種,穩(wěn)態(tài)電能質(zhì)量的各種問題基本已經(jīng)妥善解決,我們現(xiàn)在討論、亟待解決的各類電能質(zhì)量問題均指的是暫態(tài)電能質(zhì)量方面的問題,暫態(tài)電能質(zhì)量問題主要是包括電壓驟升、電壓驟降、電壓中斷、電壓暫態(tài)震蕩與諧波等在內(nèi)的一系列電壓質(zhì)量問題。如何能夠快速準(zhǔn)確的檢測擾動發(fā)生的起止時刻是暫態(tài)電能質(zhì)量檢測的主要研究工作。我們都知道,在實際應(yīng)用中,電壓信號都是帶有噪聲干擾的,那么此時如果直接對含噪信號進行擾動檢測和定位,必定會出現(xiàn)較大的誤差,如何使電壓信號噪聲降低或消失,成了檢測定位研究工作的首要任務(wù)。
信號去噪的方法有很多,包括傅里葉法、小波分析以及小波包分析,由于傅里葉變換反映的只是信號的整體特征,常用于平穩(wěn)信號的分析,我們這里討論的帶擾動的電壓信號均是暫態(tài)非平穩(wěn)信號,所以在這里我們對此方法不予考慮。……