电器与能效管理技术 ›› 2026, Vol. 0 ›› Issue (6): 28-36.doi: 10.16628/j.cnki.2095-8188.2026.06.004

• 研究与分析 • 上一篇    下一篇

基于FFT-卡尔曼滤波的谐波检测与有源电力滤波器控制研究

卢文将1, 赵昀2   

  1. 1 国网山西省电力有限公司清徐县供电分公司, 山西 太原 030499
    2 国网山西省电力有限公司太原供电分公司, 山西 太原 030021
  • 收稿日期:2025-12-16 出版日期:2026-06-30 发布日期:2026-07-14
  • 作者简介:卢文将(1993—),女,工程师,主要从事电能质量分析与谐波治理技术的研究工作。|赵昀(1991—),男,工程师,主要从事电能质量分析与谐波治理技术的研究工作。

Research on Harmonic Detection and Active Power Filter Control Based on FFT-Kalman Filter

LU Wenjiang1, ZHAO Yun2   

  1. 1 State Grid Shanxi Electric Power Co., Ltd.Qingxu County Power Supply Branch, Taiyuan 030499, China
    2 State Grid Shanxi Electric Power Co., Ltd.Taiyuan Power Supply Branch, Taiyuan 030021, China
  • Received:2025-12-16 Online:2026-06-30 Published:2026-07-14

摘要:

针对传统谐波检测方法实时性差、精度不高等问题,提出了一种基于快速傅里叶变换(FFT)与卡尔曼滤波相结合的谐波检测方法,并应用于并联有源电力滤波器(APF)控制系统。首先利用快速傅里叶变换算法对电流信号进行频谱分析,提取各次谐波分量的幅值和相位信息;然后采用卡尔曼滤波对检测结果进行动态估计和噪声抑制,提高检测精度和抗干扰能力,并采用直流侧电压比例-积分(PI)控制与电流滞环跟踪控制相结合的双闭环结构。仿真结果表明,该方法能快速、准确地检测电网谐波电流,补偿后电源电流总谐波畸变率由25.38%降至2.67%,满足国家标准要求,验证了方法的有效性。

关键词: 快速傅里叶变换, 卡尔曼滤波, 谐波检测, 有源电力滤波器, 电能质量

Abstract:

Aiming at the defects of poor real-time performance and low detection accuracy existing in traditional harmonic detection algorithms, a harmonic detection method combining fast Fourier transform (FFT) and Kalman filter is proposed, which is applied to the control system of shunt active power filter (APF). Firstly, the FFT algorithm is used to perform spectrum analysis on current signals and extract the amplitude and phase of each harmonic component. Secondly, the Kalman filter is adopted to dynamically estimate detection data and suppress noise, so as to improve detection precision and anti-interference capability. A double closed-loop control structure combining DC-side voltage PI control and current hysteresis tracking control is adopted in the system. Simulation results indicate that the proposed method can detect grid harmonic currents rapidly and accurately. After compensation, the total harmonic distortion of source current drops from 25.38% to 2.67%, which satisfies the national standard requirements and verifies the validity of the proposed method.

Key words: fast Fourier transform, Kalman filtering, harmonic detection, active power filter, power quality

中图分类号: