基于CEEMDAN的MEMS加速度计输出信号降噪方法

针对微机电系统(MEMS)加速度计输出信号存在误差,导致高压输电杆塔倾斜监测系统的输出倾角数据精确度不高的问题,提出了一种基于自适应噪声完备集合经验模态分解(CEEMDAN)联合奇异值分解(SVD)对杆塔的加速度计输出信号降噪方法。利用CEEMDAN对原始加速度计输出信号进行分解,得到一系列模态分量,分别计算其排列熵(PE),筛选出特征分量和含噪特征分量,然后再将需进一步降噪的特征分量通过SVD进行二次滤波,最后将降噪后的特征分量与未处理的特征分量进行叠加即得到降噪后的加速度计输出信号。仿真实验结果表明,所提出的方法可以有效地抑制噪声干扰,通过与扩展卡尔曼滤波和CEEMDAN-PE对比说明该方法滤波效果更好,有效提高了加速度信号分析精度和杆塔倾斜角测量精度。

Noise Reduction Method of Output Signal of MESE Accelerometer Based on CEEMDAN

In order to solve the problem that there is error in the output signal of micro electro mechanical system(MEMS) accelerometer,which causes low accuracy of the output angle data of high voltage transmission tower tilt monitoring system,a noise reduction method based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) combined with singular value decomposition (SVD) is proposed.The raw accelerometer output signal is decomposed by CEEMDAN to obtain a series of modal components,whose permutation entropies(PE) is calculated respectively,to extract components of features and components of features with noise.Then the components of features that need further to get noise reduction are carried on secondary filtering through SVD.Finally,the denoised components of features are overlapped with the unprocessed components of features to obtain the accelerometer output signal with the noise reduction.The simulation results show that the proposed method can effectively suppress noise interference.Compared with the extended Kalman filter and CEEMDAN-PE,the filtering effect of this method is better,and the accuracy of accelerometer signal analysis and tower tilt angle measurement is effectively improved.