中文摘要：

阐述了傅里叶级数法利用原始数据来重构窄带干扰的基本思想，分析了其重构精度受频率估计偏差和随机干扰影响较大的不足。在此基础上，结合遗传优化算法提出了一种抑制局部放电信号中窄带干扰的新方法。它不需要预知窄带干扰频率的精确值，而是通过遗传算法优化选择重构参数，避免了求解冗余方程组带来的累积误差，具有较强的自适应逼近能力和抗随机干扰能力，较适合实际应用。仿真和实测数据的处理结果表明：新方法提高了窄带干扰的重构精度，在含有随机干扰的情况下，依然具有较好的参数估计能力。

英文摘要：

The basic concept to reconstruct narrow-band interference by Fourier series with original data is introduced and the great influence of frequency estimation error and random interference on the reconstruction accuracy is analyzed, based on which, the partial discharge narrow-band interference elimination based on genetic algorithm and Fourier series is proposed. It doesn＇t need in advance the accurate interference frequency and the reconstruction parameters are chosen and optimized by genetic algorithm,which avoids the accumulated error caused by the solution of redundant equations. It has good adaptive approximation ability and anti-interference ability,suitable for practical applications. Simulative and experimental results show that,the proposed method improves the reconstruction accuracy of narrow-band interference and has good parameter estimation ability under random interferences.