改进扩展卡尔曼滤波方法

Improved extended Kalman Filter method

针对低通滤波的有源电力滤波器(Active Power Filter,APF)补偿精度低的问题,提出一种改进扩展卡尔曼滤波(ExtendedKalmanFilter,EKF)方法,通过提高谐波电流的滤除能力进而改善APF谐波补偿精度.对改进EKF原理进行分析,引入谐波电流代替扩展卡尔曼滤波的过程噪声,对噪声进行重构,证明噪声重构满足扩展卡尔曼滤波算法,并推导出扩展卡尔曼滤波方程.利用噪声信号统计特性,滤除特征谐波的同时对非特征谐波进行滤除,并通过无差拍预测控制削弱数字信号处理过程中的延时影响.利用Matlab搭建仿真模型与常规低通滤波器分别就滤波品质及APF补偿效果进行对比实验.研究结果表明:改进EKF滤波可有效降低电网电流畸变率,提高APF补偿精度.

Aiming at the problem that low compensation accuracy of the active power filter based on low-pass filter,this paper proposes an improved extended Kalman filter method to improve APF harmonic compensation accuracy by improving the filtering ability of harmonic current.Firstly,the improved EKF principle is analyzed,and the harmonic current instead of the process noise of the extended Kalman filter is introduced to reconstruct the noise.By proving the noise reconstruction satisfies the extended Kalman filter algorithm,and the extended Kalman filter equation is derived.Then,the characteristic harmonics are filtered and the non-characteristic harmonics are filtered at the same time based on the statistical characteristics of noise signal,and delay effect in the process of digital signal processing is weakened by deadbeat predictive control.Finally,Matlab is used to build a simulation model for comparison experiments with conventional low-pass filter in terms of filtering quality and APF compensation effect.The results show that improved EKF can effectively reduce current distortion rate of the power grid and improve the compensation accuracy of APF.