改进FFT动态信号分析方法及在电力谐波检测中的应用

项目名称： 改进FFT动态信号分析方法及在电力谐波检测中的应用

项目编号： No.60872128

项目类型： 面上项目

立项/批准年度： 2009

项目学科： 无线电电子学、电信技术

项目作者： 滕召胜

作者单位： 湖南大学

项目金额： 31万元

中文摘要： 为实现基于嵌入式系统的复杂信号动态、准确分析，本项目深入从构建具有优良旁瓣性能、可充分抑制频谱泄漏的新型窗函数入手，研究加窗插值改进FFT信号谱分析方法，提出并构建了余弦偶次幂窗、三角自卷积窗、Hanning自卷积窗、Nuttall自卷积窗等频谱函数简单、旁瓣和主瓣性能优越的新型窗函数，取得了良好的频谱泄漏抑制效果；基于新型窗函数，研究了任意谐波频谱相互泄漏干扰模型，建立了复杂动态信号频谱自适应校正方法，实现了任意次谐波的快速、准确分析和谐波间频谱泄漏分析；根据所建立的新型窗函数的频谱特征，可确定信号频谱的峰值搜索性能函数，实现信号频谱的自适应分割，提高信号参数检测的跟随性能；大量仿真实验证明了算法对复杂信号谐波分量动态分析的高准确性和快速性；构建了基于ADC+DSP+ARM的信号测量与谱分析平台，研制了一种新的谐波分析与谐波电能计量装置，实现了电力系统谐波的准确测量和谐波电能计量。项目研究形成了一套完整的基于嵌入式系统的复杂动态信号谐波分析方法，所取得的研究成果对复杂信号分析、数字信号处理、微弱信号检测等具有参考与借鉴价值。

中文关键词： 余弦偶次幂窗；自卷积窗；FFT；谱分析；谐波

英文摘要： To realize accuracy analysis of complex dynamic signal in embedded system, new kinds windows of high performance are proposed which can suppress the spectral leakage. The signal spectrum analysis method based on improved windows and interpolated FFT is studied in the project. Then, Even order cosine window,Triangular self-convolution window,Hanning self-convolution window and Nuttall self-convolution window are proposed and constructed. The spectral leakage can be more efficiently suppressed by using these new windows, which have simple spectrum and superior sidelobe & main lobe performance. The leakage and interference model with these proposed windows is researched, self-adapting correction method of complex dynamic signal is constructed, and efficient & accurate analysis of arbitrary harmonic and spectrum leakage analysis are realized. According to the characteristics of the proposed new windows, the performance function of peak value searching can be determined, self-adapting segmentation of frequency spectrum of signal can be achieved, and the tracking performance of parameter detection can be improved. The results of many simulation experiments show that the proposed algorithm is greatly accurate and efficient to the dynamic analysis of complex harmonic component. A kind of signal measurement and spectrum analysis platform composed by ADC+DSP +ARM is presented, and a new harmonic analysis and harmonic energy measurement device is Developed, which achieve accuracy measurement of harmonic and harmonic energy in power system. The research forms a complete set of complex dynamic signal harmonic analysis method based embedded system, the research achievement has a reference value for the complex signal analysis, digital signal processing and weak signal detection.

英文关键词： Even order cosine window; self-convolution window; FFT; spectrum analysis; harmonic