电工技术学报2024，Vol.39Issue(13)：4015-4024,10.DOI:10.19595/j.cnki.1000-6753.tces.230786
基于自适应移频滤波的电力系统谐波分析方法
Power System Harmonic Analysis Method Based on Adaptive Frequency-Shift Filtering
摘要
Abstract
With the increasing proportion of various types of nonlinear and impactive loads in the power grid,especially the large-scale integration of new energy sources represented by photovoltaic and wind power,the harmonic issues in the current power system are becoming increasingly severe.Accurately estimating harmonic parameters is not only critical for harmonic control but also an essential basis for power companies and users to evaluate power quality.To achieve accurate and effective detection of harmonic parameters in modern power systems,this paper proposes a power system harmonic analysis method based on adaptive frequency-shift filtering,which estimates harmonic parameters by dynamically and adaptively adjusting frequency deviations. This method first performs a frequency-shift operation on the original power grid sampled signal based on the nominal frequency ωnom.Then,considering that it is difficult to completely filter out frequency components other than the target frequency component in a single filtering operation,this paper performs multiple iterations of filtering on the frequency-shift signal.Specifically,the frequency-shift signal is convolved with a p-order self-convolution mean filter(SCMF)to reduce the interference from other frequency components on the target frequency component.After the initial frequency-shift filtering of the power grid harmonic signal,the harmonic angular frequency deviation Aω,the actual fundamental frequency fr,and the new frequency-shift function corrected by Aω can be obtained.Next,according to the desired harmonic order hs and the frequency-shift function after correction,a second frequency-shift filtering process is carried out,moving the desired harmonic spectral line to the zero frequency position.Finally,the amplitude and phase of the desired harmonic component can be accurately obtained by using the derived formulas.For a power grid sampled signal with length N and using fast convolution for the convolution operation in the algorithm,the time complexity of the proposed algorithm is O(Nlog2N).Compared with the windowed interpolation FFT algorithm(WIFFT),the time complexity of both algorithms is similar.However,the WIFFT algorithm requires prior determination of fitting coefficients through polynomial fitting.Therefore,the algorithm proposed in this paper is more convenient and straightforward in practical applications,thus offering distinct advantages. This paper conducted simulation experiments on the algorithm under conditions of weak amplitude harmonic signal,fundamental frequency deviation,noise impact,etc.,and also established an actual hardware testing platform.The amplitude relative error and the phase absolute error of the actual measurement results fully meet the requirements of Class A meters in GB/T 14549"Quality of electric energy supply—Harmonics in public supply network".Simulation experiments and practical measurement results demonstrate that the proposed method can accurately and effectively detect power system harmonic parameters,showing significant advantages in robustness,adaptability,and real-time performance.Especially for detecting weak amplitude harmonic components,this method exhibits strong applicability.The proposed method can provide new solutions and ideas for the effective and fast analysis of harmonic parameters in complex power grid environments,as well as theoretical and engineering application issues related to electric energy science and accurate metering in the electric power system.关键词
电力系统分析/电能质量/电网谐波/移频滤波Key words
Power system analysis/power quality/grid harmonics/frequency-shift filtering分类
信息技术与安全科学引用本文复制引用
李建闽,曹远远,姚文轩,梁成斌,于佳琪..基于自适应移频滤波的电力系统谐波分析方法[J].电工技术学报,2024,39(13):4015-4024,10.基金项目
国家自然科学基金(51907062,52207076)、湖南省自然科学基金(2021JJ40354)和湖南省自科区域联合基金(2023JJ50025)资助项目. （51907062,52207076）
