基于LMS自适应滤波算法的电力变压器有源降噪系统

发布时间：2018-03-28 22:30

本文选题：电力变压器噪声　切入点：有源降噪系统　出处：《高电压技术》2016年07期

【摘要】：为了有效抑制电力变压器产生的高分贝低频噪声,针对变压器有源降噪系统开展了相关研究。结合实验条件,在分析变压器噪声时域及频域特点的基础上,合理选择了有源降噪系统类型。通过仿真和理论分析了收敛系数、滤波阶数以及系统采样频率对有源降噪系统采用的LMS自适应滤波算法收敛性、稳定性以及降噪效果的影响。搭建了变压器噪声在线监测及有源降噪软件控制平台,并针对变压器噪声特点合理选择了硬件系统设备。结果表明,在变压器噪声有源降噪实验中,噪声低频区域降噪效果显著,在误差传声器处取得了5~10 d B的降噪效果,平均声能量密度下降了68.38%~90%,验证了LMS算法的有效降噪性,以及所构建的有源降噪系统对变压器噪声在线监测与抑制的可靠性。
[Abstract]:In order to effectively suppress the high decibels low frequency noise produced by power transformer, the related research is carried out for the transformer active noise reduction system. Based on the experimental conditions, the characteristics of transformer noise in time domain and frequency domain are analyzed. The type of active noise reduction system is reasonably selected. The convergence of LMS adaptive filtering algorithm for active noise reduction system is analyzed by means of simulation and theory, including convergence coefficient, filter order and system sampling frequency. The software platform for on-line monitoring of transformer noise and active noise reduction software is built, and the hardware system equipment is reasonably selected according to the characteristics of transformer noise. The results show that, In the experiment of active noise reduction of transformer noise, the noise reduction effect in the low frequency region of noise is remarkable. The noise reduction effect of 510dB is obtained at the error microphone, and the average acoustic energy density is reduced by 68.3890%, which verifies the effective noise reduction of LMS algorithm. And the reliability of the active noise reduction system for on-line monitoring and suppression of transformer noise.
【作者单位】：山东大学电气工程学院;山东省特高压输变电技术与装备重点实验室;
【基金】：国家自然科学基金(51207084) 山东省优秀中青年科学家科研奖励基金(BS2013NJ024)~~
【分类号】：TM41

【相似文献】